Caro, Dick - Wireless Networks for Industrial Automation-ISA (2008)

Wireless Networks
for
Industrial Automation
3rd Edition
by Dick Caro
Notice
The information presented in this publication is for the general
education of the reader. Because neither the author nor the publisher has
any control over the use of the information by the reader, both the
author and the publisher disclaim any and all liability of any kind
arising out of such use. The reader is expected to exercise sound
professional judgment in using any of the information presented in a
particular application.
Additionally, neither the author nor the publisher have investigated
or considered the effect of any patents on the ability of the reader to use
any of the information in a particular application. The reader is
responsible for reviewing any possible patents that may affect any
particular use of the information presented.
Any references to commercial products in the work are cited as
examples only. Neither the author nor the publisher endorses any
referenced commercial product. Any trademarks or tradenames
referenced belong to the respective owner of the mark or name. Neither
the author nor the publisher makes any representation regarding the
availability of any referenced commercial product at any time. The
manufacturer's instructions on use of any commercial product must be
followed at all times, even if in conflict with the information in this
publication.
Copyright © 2008
ISA–The Instrumentation, Systems and Automation Society
67 Alexander Drive
P.O. Box 12277
Research Triangle Park, NC 27709
All rights reserved.
No part of this work may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without the prior written
permission of the publisher.
Preface
One of the most costly items in the instrumentation and control
of any manufacturing process is the installation of connecting
wires. Indeed, much of the effort devoted to sensor networks
and fieldbuses has been justified by the reduction in cost they
provide both in terms of the initial installation of wiring and,
even more, for its long-term maintenance. Many of the faults in
wired industrial networks can be traced back to faults in wiring and connectors. For that reason, there is a very strong interest in wireless technology because it reduces the cost of
installation and maintenance. Moreover, wireless also solves
another problem encountered only in some chemical and
petroleum plants – ensuring intrinsic safety.
In many applications, wireless technology has already begun
to displace wired equivalents. In the first year of the twentyfirst century, cordless telephones first began to outsell wired
telephones. By early 2004, small office and residential networks
had become a booming market thanks entirely to the economic
and reliability advantages of wireless LANs based on Wi-Fi
technology. We are also beginning to see wireless telephones—
otherwise known as cell phones or mobile phones—displace
landline telephones. This trend has been accelerated by recent
decisions of the U.S. Federal Communications Commission
(FCC) permitting users to retain their telephone numbers when
changing wireless telephone carriers and to transfer their landline telephone numbers to cell phones.
The expectation is that the manufacturing industries will soon
adopt wireless technology. This is not the case! Industry
expects more than does the small or home office. It demands
much more. Just as there can be a two- to threefold difference
in selling price between home Wi-Fi access points and those
used for business, so too wireless on the factory floor will cost
much more to provide the reliability and performance all
industrial processes demand.
xiii
xiv
Preface
Flux is a wonderful scientific word that refers to the flow or
lines of force of an electric or magnetic field. Applied to wireless technology flux connotes the essence of change. Any
investment in wireless technology today will be obsolete in
three years. Yet, the compelling benefits of wireless are causing
homeowners and businesses to spend millions on this obsoletethe-moment-you-buy-it technology. Wireless technology is
now in a period of high flux. Wireless product life cycles have
half-lives of a few months, and suppliers must introduce new
products every month to stay competitive. Supposedly rocksolid truths are rapidly being destroyed by new technology
and new discoveries – almost daily.
Not all of the changes in the wireless market are due to technology. Many changes stem from the decisions of standards
committees, such as the approval of a new standard. Other
sources of change are government laws, the rules of regulating
commissions, and court decisions. However, the most significant sources of change are the pricing decisions made by suppliers. When the selling price of a particular technology hits a
commodity point, that technology becomes popular, driving
selling prices even lower. For example, the approval of IEEE
802.11g, often called Wireless-G, changed the Wi-Fi marketplace
in less than two months because it led to pricing points only 20
to 50 percent higher than 802.11b, the incumbent market standard before the approval. This is expected to happen all over
again when the IEEE 802.11 committee finally approves versions “n” and “s”.
Why do pricing and other decisions of the commercial marketplace affect the product technology of the industrial market?
There are two factors: volume-related cost and reliability.
When the consumer and commercial markets heat up and sales
volumes approach the millions of units, the cost for all versions
of that technology decrease for all markets. Manufacturing a
unit for the industrial market always costs more than manufacturing it for either the consumer or commercial market, but the
product cost for an industrial version will still come down
when overall sales volume goes up. This is particularly true
when the bulk of a product’s functionality resides in its semi-
Preface
xv
conductor components, since the cost of producing an industrial chip is, if anything, less than 20 percent more than that of
the corresponding commercial-grade chip. In today’s wireless
communications market, practically all product features are
implemented in VLSI (very large scale integrated) circuits,
which enable suppliers to add value in software, packaging,
and power supplies.
The reliability of almost all wireless devices at the consumer,
commercial, and industrial levels has been so good that it is
difficult to find any real differences between them except in the
area of environmental protection. Industrial products usually
need protection from high or low temperatures, high vibration,
and sometimes from chemical corrosive attack. The high reliability of consumer wireless products results from the fact that
the most critical circuitry is located in the VLSI components,
where manufacturers have gained considerable experience
producing product in high volumes. The combination of high
reliability and low purchase price has made it possible for even
consumer-grade wireless components to find application in
both commercial and industrial applications. However, the
presence of high electro-magnetic fields (EMF) in many industrial applications may often make wireless less reliable than in
the home or office. Likewise, the presence of so much heavy
steel equipment and building structures in a typical process
both blocks signals and causes reflections that interfere with
wireless signal propagation.
This book is designed to enable you to keep up with the wireless market so you can make better decisions for your products,
services, and applications. My e-mail address is provided
below to encourage you to suggest additional topics for later
editions and to correct the inevitable errors and omissions.
(Problem solving and product planning is what I do for a living, so please don’t use this e-mail address to request solutions
for your problems or that I design/specify your products.) To
suggest new topics or technologies for future editions, to report
errors and omissions, or to make any other contact related to
this book, please e-mail me at [email protected].
Table of Contents
Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Preface
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Unit 1:
Wireless Network Technology . . . . . . . . . . . . . . . . . . 1
1.1
1.2
1.3
1.4
1.5
Unit 2:
2.1
2.2
2.3
2.4
2.5
Unit 3:
3.1
3.2
3.3
3.4
3.5
Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Wi-Fi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Bluetooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Proprietary or Non-Standard Wireless Networks . . . . . . . . . . . . . 7
Wireless versus Wired Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.1 Signal Loss/Fading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.2 Multipath Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 Shared Airwaves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Antenna Technology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.4.1 Antenna Size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.4.2 Omnidirectional. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.4.3 High-Gain Directional . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.4.4 Planar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.4.5 Phased Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Wireless Network Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.5.1 Star . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.5.2 Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.5.3 Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Wireless Network Standards . . . . . . . . . . . . . . . . . . 27
Wireless Local Area Networks (WLAN) . . . . . . . . . . . . . . . . . . . . 28
2.1.1 Wi-Fi a/b/g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Wireless Personal Area Networks (WPAN) . . . . . . . . . . . . . . . . . 31
2.2.1 Bluetooth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.2.2 ZigBee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.2.3 WiMedia (IEEE 802.15.3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
WMAN, WiMAX (IEEE 802.16a) . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Wireless Telephony. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Convergence of Voice and Data Networks . . . . . . . . . . . . . . . . . . 52
Industrial Automation Requirements . . . . . . . . . . . 55
Environmental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
56
56
57
58
viii
Table of Contents
Unit 4:
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
Unit 5:
5.1
5.2
5.3
5.4
Unit 6:
6.1
6.2
6.3
Application of Wireless Networks to Industrial
Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59
Politics of Wireless . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Wi-Fi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Bluetooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
ZigBee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
ISA100 Standard for Wireless Industrial Networks. . . . . . . . . . . 65
4.5.1 ISA100.11a Technology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
WirelessHART . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.6.1 WirelessHART Technology . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Comparison: WirelessHART vs. ISA100.11a. . . . . . . . . . . . . . . . . 71
3G/4G for Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
On the Bleeding Edge . . . . . . . . . . . . . . . . . . . . . . . . .77
WiMAX (Worldwide Interoperability for
Microwave Access) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
UWB (UltraWideBand). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2.1 WiMedia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2.2 DS-UWB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Wireless Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Network Device Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.4.1 Optical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.4.2 Pneumatic Power. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.4.3 Magnetic Induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.4.4 Microwave Power Transmission. . . . . . . . . . . . . . . . . . . . . . 85
5.4.5 Conversion of Waste Energy . . . . . . . . . . . . . . . . . . . . . . . . . 86
Significant News for Wireless Networking . . . . . .87
Energy-harvesting Component Runs Wireless
Nets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Honeywell Introduces OneWireless Networks . . . . . . . . . . . . . . 87
Accutech Wireless Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . 89
Unit 7:
Recommendations for Wireless Networking . . . . .91
Unit 8:
Radio Frequency Tagging . . . . . . . . . . . . . . . . . . . . .93
8.1
8.2
8.3
Types of Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
8.1.1 RFID Passive Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
8.1.2 RFID Active Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
8.1.3 RFID Programmable Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.1.4 RF Data Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.1.5 Location Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Tag Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
8.2.1 EPC Global Gen2 tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Alternative RFID Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Table of Contents
8.4
8.5
ix
RF Database Tag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
RF Tag Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Unit 1:
Wireless Network
Technology
The changes in wireless technology for data networks over the
past five years have been more dramatic than the changes in
radio itself in the century since Guglielmo Marconi sent the
first telegraph signal across the Atlantic from Cornwall in the
U.K. to St. Johns, Newfoundland, on December 12, 1901. The
progress in commercial radio transmission from telegraphy to
voice to television was measured in decades. Commercial digital wireless transmission began in the mid-1990s when cellular
digital telephony—known as PCS for Personal Communications Service—replaced advanced mobile phone service
(AMPS), the then dominant analog voice transmission protocol. Digital wireless telephony technology was then split into
two competing technologies: time division multiple access
(TDMA) and code division multiple access (CDMA). TDMA is
still used by AT&T but is being phased out in favor of global
system mobile (GSM), a standard version of TDMA used by
most European and Asian carriers as well as by T-Mobile and
AT&T. CDMA is used by some Japanese carriers as well as by
Sprint and Verizon. TDMA, GSM, and CDMA are not interoperable.
The wireless local area network (LAN) began to emerge in the
late 1990s, when it became obvious that there was a need for
wireless data networking. Wireless LANs required faster data
transmission than was possible with cellular PCS (of any technology), and eventually industry settled upon using digital
spread spectrum as defined by the IEEE 802.11 standards.
Spread spectrum was originally developed for the U.S. military
so wireless transmissions could be made in the presence of
strong jamming signals. This work by the military was based
on the spread spectrum patent US 2,292,387, which had origi-
1
2
Unit 1: Wireless Network Technology
nally been granted to Hollywood actress Hedy Lamarr and her
partner George Antheil.
Frequency hopping spread spectrum (FHSS) and direct
sequence spread spectrum (DSSS), both operating up to 2.0
Mbps, were the first two IEEE 802.11 technologies. Neither is
commercially available today. These initial technologies were
improved upon until, in rapid succession, IEEE 802.11b (operating at up to 11 Mbps) and 802.11a and 802.11g (both operating up to 54 Mbps) emerged. All of these are called Wi-Fi
(wireless fidelity) after the name of the supporting industry
association, the Wi-Fi Alliance, but 802.11b has become a commercially successful technology with a large installed base.
Both 802.11a and 802.11g are rapidly penetrating the market,
essentially displacing 802.11b. For the sake of simplicity, I will
continue using 802.11a, 802.11b, and 802.11g to designate each
of the IEEE 802 standards, but the marketplace calls these technologies Wireless-A, Wireless-B, and Wireless-G, respectively.
1.1
Standards
The dynamic nature of wireless digital data communications
stems from the standards committees of the Institute of Electrical and Electronic Engineers (IEEE), which develops most of
these protocols. No illusion currently exists in the IEEE 802
committee that is responsible for personal area network (PAN),
LAN (local area network), MAN (metropolitan area network),
and WAN (wide area network) that it will be possible to create
a single network protocol useful over all of these four domains.
Therefore, each application that has a special interest that is not
accommodated by an existing protocol can form a new subcommittee to create a new protocol. IEEE ensures only that
these committees deliberate fairly, do not exclude a genuine
interest, and that all proposed standards are publicly reviewed.
All IEEE 802 standards are automatically submitted to the
ISO/IEC (International Organization for Standardization and
International Electrotechnical Commission) for consideration
as international standards. Several of the IEEE 802 standards
have failed in the commercial marketplace, while others have
succeeded.
Unit 1: Wireless Network Technology
3
Another source of wireless communications protocols is the
International Telecommunications Union (ITU), the standards
body for telephone networks. With the conversion of telephony from a purely wired circuit-switched analog service—
also known as POTS (plain old telephone service) – to 3G
(third-generation) wireless, a technology convergence is underway in data networks. 3G is a wireless packet-switched digital
service based on CDMA, and is an ITU standard now being
commercialized in several countries with worldwide adoption
was scheduled by 2005, but continued popularity of GSM has
delayed implementation in most countries. In fact, many have
predicted that home and mobile computing will soon use
broadband wireless packet switching rather than either telephone DSL (digital subscriber loop) or CATV (community
antenna television) cable modem services. Even though there
are excellent reasons to keep 3G wireless in mind for in-plant
voice networks and some mobile data applications, it is not
presently being considered for industrial use. However, given
the eventual availability of low-cost and low-power-consumption 3G, and the likely successor, 4G, it cannot be ignored.
A word of caution is in order about the standards documents
for data communications. These very large documents are not
intended to be read by users. They are written for the implementer of networks and networking devices. If you really want
to see some examples of such standards, however, most IEEE
802 network documents more than six months old are available
for download on the IEEE standards web site:
http://standards.ieee.org/getieee802/ (look for the click here
link in the last paragraph of the text). The IEEE and others
often publish books about the standards, making them easier
to understand.
1.1.1
Wi-Fi
One factor causing rapid technological change in wireless communications is the ever-increasing capacity of commercial
semiconductor processes such as CMOS (complementary
metal oxide semiconductor) to handle higher frequencies. This
factor alone is responsible for the recent rise in interest in
4
Unit 1: Wireless Network Technology
802.11a, which previously required more expensive GaAs (gallium arsenide) processes or higher-power bipolar semiconductors. When 802.11a parts are built in CMOS, they are as
economical as the slower 802.11b parts.
With the ratification of 802.11g and the subsequent flood of
new products on the market, we are now witnessing another
dramatic change in the Wi-Fi market. It seems certain that by
mid-2004 802.11g, which is backwards-compatible with
802.11b, has completely displaced 802.11b in the new wireless
products marketplace. Since 802.11a and 802.11g share a common modulation technology—namely, orthogonal frequency
division multiplexing (OFDM)—products that offer both standards are not only possible but also economical. As the Wi-Fi
band at 2.4 GHz becomes saturated, the benefits of 802.11a
become compelling since the 5.0 GHz band currently used for
802.11a offers eleven non-overlapping channels, versus three
for 802.11b and 802.11g. A recent FCC regulation makes thirteen additional channels available for 802.11a, for a total of
twenty-four non-overlapping channels. Chips that offer
802.11a/b/g are already on the market, and soon all new Wi-Fi
LANs will offer all three technologies at little to no price premium. Figure 1 illustrates a roadmap for this transition in the
Wi-Fi market.
IEEE 802.11n is nearing completion as a standard; however, the
specification in draft form has long been available. Many suppliers have now released “pre-n” products promising to
upgrade them when the final standard is finally approved. The
most appealing technology embedded into 802.11n is called
MIMO (Multiple Inputs, Multiple Outputs) most easily recognized by several antennas on 802.11n products. The “n” standard calls for all signals to be simultaneously transmitted on
each of the antennas. Due to the spatial diversity of these
antennas (they are a few centimeters apart,) signals transmitted
by all antennas will be received by the multiple antennas of the
receiver slightly out of phase with each other. MIMO technology provides a way for the receiver to align the phases of the
received signals such that the resulting resolved signal is now
stronger and more reliable than any single signal. Note also
Unit 1: Wireless Network Technology
5
Roadmap for Wi-Fi
802.11n
Market Size
802.11b/g
802.11g
802.11b
2000
2001
2002
2003
802.11a
2004
2005
2006
2007
2008
2009
2010
CMC Associates Estimate, 2007
Figure 1. Roadmap for Wi-Fi
that reflected signals, often called multipath signals, are also
out of phase with the original. MIMO offers a technical solution to the multipath “problem” often associated with networks built in the “canyons of steel” that often describe large
plant units in the process and metals industries.
Additionally, IEEE 802.11n also bonds channels in both the 2.4
and 5 GHz ISM bands that were formerly assigned to 802.11g
and 802.11a. This means that currently, an IEEE 802.11n device
requires two radios, one for each band. In the future, a single
software-defined radio may be able to solve this same problem.
Many inexpensive “pre-n” devices may not be able to implement the dual radio part of IEEE 802.11n simply because they
do not have a 5 GHz radio.
Channel bonding in 802.11n may be used to achieve a higher
data rate. While a single channel for either a or g can achieve a
theoretical 54 Mbps, bonding two channels can achieve a theo-
6
Unit 1: Wireless Network Technology
retical 108 Mbps. 802.11n can achieve data rates as high as 480
Mbps by bonding nine channels.
The Wi-Fi market is supported by The Wi-Fi Alliance, which in
its own words is “is a nonprofit international association formed in
1999 to certify interoperability of wireless Local Area Network products based on IEEE 802.11 specification. Currently the Wi-Fi Alliance has over 200 member companies from around the world, and
over 1000 products have received Wi-Fi® certification since certification began in March of 2000. The goal of the Wi-Fi Alliance's members is to enhance the user experience through product
interoperability.” The Wi-Fi Alliance website is http://
www.weca.net/OpenSection/index.asp.
1.1.2
Bluetooth
Bluetooth has already been applied in many commercial products but at a much slower pace than its developers ever
dreamed. Originally defined to replace wire/cable technology
for cellular telephony, such as for connecting headsets, it had
just enough networking capability to interest a wide variety of
companies to extend its use beyond its original scope. In fact,
Bluetooth is far more than a communications protocol; it is a
full communications application stack. The lower two communications layers of Bluetooth (PHY and MAC) have been published as the IEEE standard 802.15.1. For the original task of
device connection, Bluetooth offers a rich suite of functionalities, including enabling walk-up linking without user interaction and establishing voice connection. Bluetooth networking
is intentionally limited to a maximum of eight Bluetooth nodes,
which together form a piconet. When a node is included in
more than one piconet, that node then assumes the routing task
of forwarding messages to/from the other piconet, adding a
form of mesh networking to the complexity of Bluetooth. The
most attractive feature of Bluetooth for industrial automation
purposes is its use of forward error correction (FEC) for delivering messages without error and without requiring retransmission. The drawback of FEC is loss of efficiency: a 1 Mbps
communications channel can deliver only 721 Kbps.
Unit 1: Wireless Network Technology
7
A multivendor consortium defined Bluetooth, not a standards
organization. With consent of the Bluetooth Alliance, the lower
two layers of Bluetooth were reformatted and have now
become the IEEE 802.15.1 standard. Just like 802.11b and
802.11g, it operates in the unlicensed 2.4 GHz frequency band,
but uses frequency-hopping spread-spectrum technology that
hops faster than the original FHSS of 802.11. As a result, the
presence of Bluetooth in close proximity to Wi-Fi nodes causes
the signal for the WLAN to degrade, sometimes spelling disaster for Wi-Fi transmissions. Bluetooth 1.2 and later protocols
help such nodes avoid signal degradation by listening for signals on the radio channels before transmitting. Many early suppliers of nodes with both Bluetooth and Wi-Fi have been able
to synchronize transmissions to avoid degradation. Suppliers
of 802.11a, which operates in the 5 GHz unlicensed band, are
quick to point out that they avoid signal degradation from
Bluetooth completely. Nevertheless, 802.11g suffers the same
potential problems as 802.11b in the presence of Bluetooth.
If you want to know more about Bluetooth, a rich source of
information can be found on the official Bluetooth SIG website:
http://www.bluetooth.com/help/. If you want to develop
Bluetooth products, the Bluetooth developers’ website offers
lots of reference material and discussion groups: https://
www.bluetooth.org/.
1.2
Proprietary or Non-Standard Wireless Networks
Standards take a long time to be developed, much slower than
the pace of technology. Commercial suppliers often cannot
wait for the approval of a standard, or may have a product concept that adequately fulfills the network requirements more
than any proposed standard. These companies will often introduce their network products hoping to establish a market in
the absence of standardized networks. The experience gained
by these suppliers can often be highly useful to the designers of
network standards. Sometimes, this network can become a
standard.
8
Unit 1: Wireless Network Technology
Currently, two suppliers, Honeywell and Adaptive Instruments, both offer their own wireless networks for process control field instrumentation. Both networks use frequency
hopping spread spectrum operating in the 915 MHz ISM
(Industrial, Scientific, and Medical) band. These networks are
capable of passing data at rates that vary from 4.8 to 76.8 Kbps
over distances that vary from 780m to 175m respectively. Their
devices are battery powered and have battery life estimated to
be several years. Both of these networks are configured with a
wired base-station located close to the field instruments, and
form direct links to each instrument from the base station.
Additionally, Dust Networks is another supplier using frequency hopping in the 915 MHz ISM band, but with integral
mesh networking technology. Dust sells OEM modules to be
used by other manufacturers to build wireless transmitters.
Emerson Process Management is using Dust Networks technology for its first generation wireless instrumentation.
1.3
Wireless versus Wired Networks
Wi-Fi has generally been considered to be Wireless Ethernet,
but it is far more than that. Wired networks, such as Ethernet,
are designed for communications between fixed locations.
Wireless networks, such as Wi-Fi, are designed for communications between devices. The distinction is lost for fixed-location devices, but device mobility is a primary benefit of
wireless. Mobile applications are often found in discrete parts
manufacturing and assembly and in all types of warehouse
applications. However, the primary applications for wireless in
industrial automation are expected to be between fixed locations.
The air is free, but to operate, wireless networks will often need
a wired connection to a computer or to the wired network, a
source of power, and radios. Estimating the cost of a wired network is easy. It is the sum of the cost of the network cable, junctions, and connecting wires; the cable and junction installation;
the network interfaces; and the long-term maintenance of the
installed wiring plant. Wireless networks are more difficult to
Unit 1: Wireless Network Technology
9
estimate. They include the cost of wiring to access points,
access point equipment, wireless interfaces, and long-term
wireless troubleshooting and maintenance. While there are
fewer items to install and maintain, experience with installation and maintenance of wireless equipment is much more limited than wired.
The other notable problem of wireless devices is that they still
need a power source. Wired network nodes can draw power
from the local AC receptacle, but mobile wireless devices
depend on batteries or some alternative power source. Of
course, you can always plug the wireless device into a local
power source, but then you lose the mobility advantage and
incur the cost of installing power connections at the device. To
some extent, the recent PoE (Power over Ethernet) standard,
IEEE 802.3af was created to help resolve part of this problem
by supplying electrical power on the wired Ethernet network
so it can be used by wireless access points. This standard seems
to be well accepted for business or commercial access points.
However, PoE still does not address the issue of powering the
wireless end-device itself.
1.3.1
Signal Loss/Fading
In the early twenty-first century, wireless networks still suffer
from mysterious dead spots – an area in which there is no
reception. We say mysterious because even very careful planning cannot remove all dead spots, and sometimes live spots
just move or, in the language of radio, fade. The spontaneous
loss of communications for no apparent reason is probably one
of the most irritating aspects of wireless. Often, the signal mysteriously returns even before the cause of its loss can be investigated. This occurs with cellular telephones, with Wi-Fi devices,
and with all other wireless LAN technologies.
Signal loss can be caused by interference from other radio signals present in the same part of the spectrum as well as by
moving equipment. Sometimes, a live spot exists only as a
result of a multipath effect when the signal is reflected from
some stationary object; sometimes the multipath signal inter-
10
Unit 1: Wireless Network Technology
feres with and cancels the primary signal causing a dead spot.
Wi-Fi seems to fade in areas in which microwave kitchen appliances are in use or in which a cordless telephone is operating at
2.4 GHz. Actually, the signal loss is due to interference that is
difficult to tell from fading.
Dead spots may occur within buildings depending on their
materials of construction. In the line of sight between the access
point and the wireless device, each time the radio wave passes
through a solid the signal is attenuated. Denser materials attenuate more than less dense materials. Metals, particularly steel,
used in building construction may absorb or attenuate most of
a radio signal, creating a dead spot in its radio “shadow.” Moving the access point or the device by a small amount, perhaps
only a few millimeters, may eliminate the dead spot.
Finally, there are sunspots! The sun emits a broad spectrum of
electromagnetic waves at all frequencies, which generally constitutes noise. Once in a while, the surface of the sun experiences flares or dark spots that emit very strong electromagnetic
waves that are known to interfere with radio transmissions,
and occasionally with wired communications as well.
1.3.2
Multipath Distortion
Radio waves move from an omnidirectional antenna in all
directions. When these radio waves strike a very dense object
such as metal or stone they are reflected, much as light is
reflected from a mirror or other shiny surface. Even when there
is a clear path between the transmitting and receiving antennas, some of the signal reflected from other paths will arrive at
the receiving antenna. This phenomenon is called multipath
and can distort the received signal since the longer path will
cause the signal to be received out of phase with the signal
from the direct path. The effect of multipath distortion can
range from nothing to the cancellation of the signal, depending
on the paths and the resulting delays. In some cases, the multipath effect can even boost the received signal. This occurs
when both paths are received in phase, such as when multiple
transmitting antennas are used. In fact, this phenomenon is
Unit 1: Wireless Network Technology
11
used by IEEE 802.11n. The technology for using the multipath
signal to enhance performance is called MIMO (Multiple Input,
Multiple Output.) MIMO uses multiple antennas on both the
transmitter and the receiver to achieve multiple transmissions,
and to receive the signals.
1.3.3
Shared Airwaves
One of the problems of radio is that the spectrum is limited,
and new uses are constantly being found for it. The attempt to
allocate certain frequency bands for specific uses is the responsibility of governmental agencies – the FCC in the United
States. The frequency assignment process is highly political
and is based loosely on technology. Furthermore, frequency
assignment is highly dynamic and sensitive to economic conditions and the appearance of new technology solutions. For
example, the FCC originally assigned eighty-two 6 MHz frequency channels exclusively for broadcast television – an enormous segment of the spectrum for a single purpose. In most
areas of the United States, only a tiny fraction of that spectrum
is actually being used in any one location, since commercial
television was reluctant to use the higher UHF frequencies
because of their limited distance reception range. Television
channels are also reused based on geography – when stations
are far enough apart to not interfere with each other. Some of
the unused UHF television channels have already been
reclaimed for other uses, and more are scheduled to be
reclaimed in the future. Needless to say, television stations are
highly reluctant to change frequency channels once they are in
use.
The U.S. military is one of the most demanding users of radio
frequencies and is very reluctant to give up any frequency previously assigned to it. This same attitude is reflected in the military establishments in most other countries as well, even when
the service using that frequency has been abandoned. Another
demanding public sector is amateur radio, which has allocated
to it small frequency bands scattered throughout the spectrum.
Amateur radio broadcasters are also reluctant to abandon any
frequency band.
12
Unit 1: Wireless Network Technology
Nevertheless, the United States and most other governments
have ordered that all allocated users share the radio spectrum
unless the service cannot function when shared. By definition,
the military frequency bands cannot be shared. Public radio,
television, and global positioning satellite (GPS) frequencies
also cannot be shared. Certain public safety and many business
uses are licensed and are not shared. The remainder can be
shared, and they are divided into both licensed and unlicensed
frequency bands. Generally, licensed bands allow users to
broadcast at higher power ratings in order to reach longer distances, while unlicensed bands are forced to limit radiated
power to minimize interference between users.
Users of shared radio frequencies demand some type of access
controls so they can avoid interference. Fortunately, as the
demands on radio bands have increased, so has the ability to
economically use higher frequencies. Expansion to higher frequencies has enabled higher rates of information exchange. But
this often results in messages of shorter length, and usually
requires sacrificing range or distance between sender and
receiver. Higher frequencies are usually limited to line of sight
between transmitter and receiver. Most of the new methods for
sharing radio frequencies have depended upon packet radio
technology that is suitable only for digital data transmissions.
In one such packet radio technology, wireless LAN, many
users may share the same frequency through the use of spread
spectrum technology. GSM is a wireless telephony standard
that is used in most of the world. In the United States it shares a
pair of frequency bands with both time division and frequency
division multiplexing. Advocates of CDMA claim it to be the
wireless telephone technology of the future, and it depends
upon packet-switching technology to share the bandwidth.
Loss of Privacy
Once a radio broadcast enters the air, or ether, as it is sometimes called, anyone may receive the signal. Wired communications require a physical electrical connection, or at least an
inductive coupling that is very close to the wire so as to intercept the signal. Governments have declared that intercepting a
Unit 1: Wireless Network Technology
13
wired communication signal is illegal and may only be permitted with a court order. No such limitations exist for most types
of radio signals. If you broadcast, anyone can receive. However, the law in the U.S. has made listening to some radio
broadcasts illegal, even though that is difficult to enforce.
Solutions exist for making radio signals more private. Though
no way exists to provide exactly the same level of privacy of an
ordinary wired communication, many methods are available
for making radio transmissions difficult to interpret, even if we
cannot make them impossible to receive. One of the most common ways to achieve privacy is to use highly directional radio
antennas in which interception would only be possible if one
had exact knowledge of and access to the line of sight between
sending and receiving antennas. Locating these line-of-sight
antennas on towers and rooftops physically limits the potential
for interception.
Using encryption can make even an intercepted signal difficult
or impossible to interpret, hopefully to the equivalent degree
as wired communications. Encryption is the science of scrambling the data using a method and a key. Decryption is the
method of using a key to unscramble the data to restore it to its
original form. The interceptor would need the encryption key
to unlock the data and decrypt it, provided that the encryption
method is known. Simple encryption is sufficient to protect
non-critical or non-vital data, but more complex encryption is
required for data exchanges that may involve personal or
financial data. Transmissions of data necessary to operate a
manufacturing production facility are considered to require
high immunity from interference or interception.
There are two types of encryption: secret or private key and
public/private key. Secret key encryption uses a key or cipher
consisting of several characters to process the original message
using a known method so as to create an encrypted message.
The same key is used to decrypt the message after it is
received. Many methods called processes or algorithms are
used for secret key encryption. The best-known algorithm is
the Data Encryption Standard (DES). It was developed by the
National Institute of Science and Technology (NIST), and is
14
Unit 1: Wireless Network Technology
widely published. DES uses a 56-bit secret key. To make it
more secure, Triple-DES is sometimes used in which the same
key is processed three times, though the key length is the same.
Advanced Encryption Standard (AES) is the latest NIST development for assuring maximum security of the secret key
method. It uses 128-, 192- and 256-bit keys.
One of the most secure methods for data transmission privacy
is the public/private key encryption method, which is used to
verify signatures. A user is given a public key that may be published. When the sender “signs” a document, the digital signature is encrypted with the sender’s private key. The encrypted
signature and the sender’s public key are both sent to the recipient, who then uses the sender’s public key to verify the signature of the original user. Document privacy is obtained by
encrypting the whole document using the recipient’s public
key. When received, the targeted recipient, and only that recipient, may decrypt the document using his or her own private
key. While complicated, no method provides greater assurance
of privacy than public/private key encryption. For public/private key systems to work effectively there must be an open
repository for public keys, such as http://www.keyserver.net/
en/ or http://pgp.dtype.org/, both of which only support
PGP (“Pretty Good Privacy”) encryption keys. Clearly, there
should be only a single key server, but this has not yet happened.
There are two dominant public/private key encryption methods: RSA (Rivest-Shamir-Adleman), and PGP (Pretty Good Privacy). RSA is a product of RSA Security, a company that
specializes in security issues. PGP is an open algorithm supported by software from PGP Corporation. Both methods can
be used, but PGP is more often used to encrypt an entire message. Secure socket layer (SSL) is the leading security protocol
on the Internet and uses RSA encryption. When an SSL session
is started, the server sends its public key to the browser. The
browser then uses the public key to send a randomly generated
secret key back to the server in order to have a secret key
exchange for that session. The problem is that the public key
infrastructure (PKI) requires too much computational logic to
Unit 1: Wireless Network Technology
15
be implemented easily on today’s very simple or battery operated devices. The use of encryption is usually limited to verifying digital signatures and to financial transactions such as a
credit card or bank account number.
Network Membership
Membership in a wired network is achieved by establishing a
physical connection to the wiring or to a network element such
as a wiring hub or switch. Wireless units are neither connected
nor disconnected from a network. In order to communicate,
they must first seek to join the wireless network. As part of the
protocol for joining the network, a network address is
assigned.
Network membership is actually a function of the routing
capability, which is embedded into all IEEE 802-based networks by using an IEEE 802.1d protocol implemented by network switches (wired) or access points (wireless.) The
algorithm is called a spanning tree bridge. In it, the network
switch or access point learns the address of each connected station when a message is sent from that station, since the from
address is located in the message header. In this way, messages
not intended for the network members for that switch or access
point do not clutter the network. For a station to join the
switch’s local membership list, it must only send a message.
Roaming is an essential property of wireless networks,
although the need for roaming exists anytime a portable computer is used on different network segments. Any wireless
device may physically be moved so as to be in the range of different wireless networks. The ability to roam means that applications may continue to perform their network
communications as the device is moved from one wireless network (domain) to another. Networks that support roaming
transfer membership transparently from one domain to
another.
For a wireless telephony network, roaming is transparent as
cell phones move from the range of one cell tower to the next.
It’s not that simple for mobile computers on a wireless LAN,
16
Unit 1: Wireless Network Technology
however. Usually, all of the wireless LAN access points are
connected to ports on a single network switch, which performs
the routing function. However, this results in a clutter of messages being sent to all access points in hopes of finding the targeted station, if it is not on the local membership list of the
network switch.
The newest solution for roaming wireless LAN stations is the
so-called wireless switch. This is an access point that has the
ability to perform advanced 802.1d spanning tree bridging
logic. Just like a wired switch, it learns that a station is within
its range when the station transmits a message. The problem is
that the station may have previously been in the range of a different access point. Recent advances to the IEEE 802.1d standard provide the network management capability to rapidly
move the registration of a station from one switch to another.
The wireless switch uses this capability to move the registration of a station from one access point/wireless switch to
another. By using wireless switches, the broadcast network
clutter is reduced.
1.4
Antenna Technology
Though the antenna is usually a passive (no electronics) element of the wireless network, it is critically important. The
antenna(s) on the transmitter couple the signal to the
antenna(s) on the receiver just as surely as wire connects wired
networks. Furthermore, just as wired network capacity tends
to relate to wire size; wireless network throughput depends
strongly on antenna gain.
One very important characteristic of antennas for wireless is
that they tend to polarize the transmitted signal. A vertical
transmitting antenna will cause a vertical polarization, while a
horizontal antenna will cause a horizontal polarization. AM
radio is vertically polarized, which is why automobile antennas are vertical. FM broadcast radio is actually horizontally
polarized much like VHF and UHF television. Using the same
vertical antenna used for AM radio to receive FM radio is a
suboptimal method. However, it still works because polariza-
Unit 1: Wireless Network Technology
17
tion is never purely horizontal or vertical, and the antennas are
also usually not exactly vertical either.
1.4.1
Antenna Size
As the frequency of the radio band has increased, antennas
have become shorter. At the common 2.4 GHz band used for
wireless LANs, a full wavelength antenna is only about 12.5 cm
(5 inches). At these sizes, it becomes possible to integrate
antennas entirely inside the product such as an instrument or a
notebook computer. The length of the antenna is optimal when
it is exactly one wavelength long. When citizens band radio at
27 MHz was popular, optimal antennas were over 11m in
length. Even automobile antennas at ¼ wavelength were
almost 3m (9 feet) long. The formula for calculating full-wavelength antenna length is the following:
wavelength(m) = 299,792,458 ÷ frequency(Hz)
1.4.2
Omnidirectional
Most antennas for wireless networks are omnidirectional –
they radiate the signal in all directions at the same time. Omnidirectional antennas are the base-case and are considered to
have zero gain measured in dB. Omnidirectional antennas are
usually polarized vertically for convenience.
Since the transmitted energy from an omnidirectional antenna
radiates equally in all directions, it effectively loses power proportionally to the square of the distance traveled. Additionally,
as radio waves pass through any matter they become attenuated in proportion to the density of that matter. In particular,
conductive metals such as copper, aluminum, iron, and steel
tend to conduct radio energy toward a grounding point, if one
exists. Finally, radio waves are received at the targeted radio’s
antenna, where that same antenna will also receive radio
energy from other transmitters and even the original signal
reflected from an obstruction in the direct path.
18
1.4.3
Unit 1: Wireless Network Technology
High-Gain Directional
To overcome the signal losses caused by omnidirectional
antennas, directional transmitting antennas concentrate the
radiated energy into a narrow beam. Directional receiving
antennas capture signals in the near vicinity of the primary
receiving antenna and reflect that energy to the primary
antenna, thus effectively increasing the gain of the receiving
antenna. Several methods are available for concentrating the
radio energy, all of which involve reflecting energy emitted in
the wrong direction and redirecting it to the target direction.
Directional antennas are intended to be aimed manually so the
receiving antenna is receiving in the direction of the transmitting antenna. If either the transmitter or receiver is in motion,
the antennas must be continuously repositioned to align.
Directional antennas also tend to eliminate noise, other stray
signals on the same frequency, and reflections, all of which
improves the signal-to-noise ratio, thus improving reception.
Directional antennas used for ultra-high-frequency (UHF) and
microwave radio are illustrated in Figures 2 and 3. Note that it
is not necessary to use a high-gain antenna for both transmit
and receive, nor is it necessary to use the same type of highgain antenna for both transmit and receive. Obviously, to realize the benefits of high-gain antennas in bidirectional service,
such as a WLAN, high-gain antennas should be used on both
ends to improve service.
The stacked YAGI antenna shown in Figure 2 illustrates a vertically polarized series of antennas. Each element tends to radiate in all horizontal directions, but not vertically. The longer
vertical elements behind those in the front are designed to
reflect the radiation that is toward the rear, back toward the
front. On reception, the rear elements tend to reflect received
signals toward the active front antenna elements. This design is
generally thought to increase the gain by 3 to 13 dB, depending
upon the number of vertical elements.
A parabolic dish, shown in Figure 3, is much more expensive
than an omnidirectional or stacked YAGI antenna, but it provides far more gain. The parabolic dish reflects the radiated
Unit 1: Wireless Network Technology
19
Figure 2. Stacked YAGI Antenna
wave into a narrow beam, and likewise focuses the received
energy from a wider area into the receiving antenna. In addition to the higher cost of the antenna and its vulnerability to
wind and snow, the parabolic dish suffers from another drawback: the narrow beam width when it is used as a transmitting
antenna. The narrow beam width makes the task of aiming the
receiving antenna much more difficult, especially when used
with a parabolic receiving antenna. A parabolic dish is generally considered to have a gain of 20 to 24 dB over an omnidirectional antenna.
1.4.4
Planar
The planar antenna design evolved from use in mobile telephones and has become available for commercial use. Planar
antennas are small and lightweight, and at wireless LAN frequencies can be embedded into the equipment. Antenna gain
can be obtained by building more than one planar antenna into
a device. Generally, the planar antenna is omnidirectional.
20
Unit 1: Wireless Network Technology
Figure 3. Parabolic Dish Antenna
1.4.5
Phased Array
A phased-array antenna is a two-dimensional organization of
planar antennas. Military radar systems were the first to use
phased-array antennas. While they will certainly be used in
commercial and industrial applications, currently their high
cost makes them unattractive. The appeal of the phased-array
antenna is that it can exhibit the high gain of a directional
antenna and can be aimed electronically without moving the
base antenna. Therefore, phased-array antennas enjoy a true
advantage when connecting wireless radios in mobile
equipment as they move beyond the range of omnidirectional
antennas.
Unit 1: Wireless Network Technology
21
Phased-array antennas form a beam electronically rather than
by using the reflective properties of metals. Each component
planar antenna must be separately driven, with the same signal
modified in phase to form this beam. Military phased-array
antennas use hundreds of elements, but when this technology
becomes commercial, many fewer elements will be necessary
to achieve a formed beam that is sufficient for industrial distances. Beam formation can usually be directed to an included
angle of about 75 degrees.
1.5
Wireless Network Topologies
Wired networks have a layout or topology that is determined
by the location of the nodes and network components. Wireless
networks are not so easily described. The topology of a wireless network is determined by the logical capabilities of the network components. Often the user must determine how the
wireless network’s topology is to be configured after installation, or perhaps after some usage determinations.
1.5.1
Star
The most typical or default arrangement for a wireless network
is a star cluster in which the wireless access point is at the center, as illustrated in Figure 4. Each wireless device then communicates only with the common access point, which is
usually connected via wires to a network switch. This arrangement then places all of the wireless devices into the same collision domain, presuming that this is an Ethernet-based
network. Usually, this arrangement presents no problem since
the access point itself will be unable to receive more than one
message at a time and will ignore whichever began second.
Unfortunately, the second device will not be notified that its
message was not completed since the message rejection
occurred at layer 1 of the network and no defined network protocol exists for layer 1 message rejection as there is for Ethernet
at layer 2. Since many messages are sent using TCP/IP protocol at layers 3 and 4 of the network, the second device will
receive notification from TCP that the message was not
acknowledged and then will be retried. If the message is sent
22
Unit 1: Wireless Network Technology
) ) ) ) )
using user datagram protocol (UDP), no such acknowledgement is provided. Rather, it must then be provided by the
application or the application layer protocol.
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
) ) )
) )
)
)
)
STAR
Figure 4. Wireless Star Topology
Wireless access point switches are now appearing for commercial networks. Their function is similar to that of an unswitched
access point, except that they carry a full layer 2 switching
function using the spanning tree bridge protocol, IEEE 802.1d
and IEEE 802.1w, and rapid reconfiguration protocol needed
for wireless roaming. Spanning tree bridge protocol allows a
network switch to learn the address of its connected devices by
listening to messages they send. It then routes any messages
received at the switch to the device, and to no other. The roaming extension allows a network-connected station to retain network sign-in while moving from the radio zone of one wireless
network switch to another. Under 802.1d, when a mobile station moved out of the range of one switch/access point, a timeout period would be necessary before that station could log
into another switch/access point. With 802.1w, the station may
log into the network at any switch/access point by just sending
Unit 1: Wireless Network Technology
23
a message, which will cause it to be logged out of the previous
switch/access point without needing the timeout period.
The significance of rapid reconfiguration for industrial automation is obvious in the case of mobile devices such as automated guided vehicles. However, rapid reconfiguration can
also be used to increase the reliability of star networks through
redundancy to configure the highly reliable networks needed
for the manufacturing environment. When applied to stationary equipment, a wireless network connection is normally
highly reliable. Due to interference in the radio spectrum, however, it is possible that messages will not reach the desired destination. In that case, a second switch/access point can provide
the redundancy needed for the alternate path required for a
highly reliable network. However, in the case of wireless networks, full 100 percent redundancy is not required. Only a viable alternate path that can serve many primary paths can serve
as a backup path.
1.5.2
Tree
As in wired networks, wireless networks can be organized into
a tree topology. Each field unit is configured to a network that
is connected to a specific switch/access point. That access point
is then hierarchically connected to another access point closer
to the wired network. The topology appears as illustrated in
Figure 5.
1.5.3
Mesh
The newest and most revolutionary form of network is called a
mesh. In a mesh network each station is both an end device
and a network forwarding element. Mesh networks are naturally self-healing and redundant – exactly the properties
needed for industrial automation networks.
In a mesh network, each station is responsible for forwarding a
network transmission not intended for itself to other stations
within its radio range. Those stations, in turn, send the transmission to at least one other station within its radio range, as
24
Unit 1: Wireless Network Technology
) ) ) ) ) ) ) )
) )
) )
)
) )
) )
) )
) )
) )
)
)
)
)
)
) )
) )
) )
)
) )
) )
) )
)
)
)
) )
) )
)
)
)
)
) )
) )
)
TREE
Figure 5. Wireless Tree Topology
illustrated in Figure 6. Therefore, the network becomes very
redundant, fault-tolerant, and extended in range. The drawback is that each station must remove redundant messages. In
effect, each mesh network station becomes a network router.
Additionally, since multiple paths are involved, each receiving
station must reject duplicate messages received from divergent
paths.
Standardized mesh network protocols also include the capability to build and maintain routing tables so as to provide clues
for forwarding messages. This prevents messages from looping
in directions other than toward their intended destination,
which results in greater network efficiency. Routing tables are
dynamically constructed as messages pass through each routing node of the mesh network. Since mesh networks that are
intended for industrial automation tend to have 256 or fewer
nodes, routing tables can be small and the routing simple.
Routing tables need to be updated when new nodes appear in
Unit 1: Wireless Network Technology
25
)
) ) )
) )
)
) )
) )
) )
)
)
)
) )
) )
) )
) )
) )
)
)
)
)
)
)
)
) )
) )
)
) )
) )
) )
) )
) )
)
) )
)
)
) ) )
) )
)
)
MESH
Figure 6. Wireless Mesh Network
the mesh or for any reason fail to respond to forwarded messages.
Mesh networks are not new. The Internet itself is a very large
wired mesh network with very complex routing algorithms.
Since IP addresses do not imply anything about location, messages routed on the Internet “hop” from one node to another
that is (hopefully) closer to the desired destination. Internet
routing algorithms are typically efficient enough that few messages need more than fifteen hops to reach their desired destination.
Wireless mesh networks pose a problem that is not encountered with wired mesh networks such as the Internet. With
wireless mesh networks, there is no way, other than by using a
highly directional antenna, to prevent a message transmitted
by one node of a wireless mesh network from being received
by other nodes. This leads to multipath routing, or message
duplication. Typically, the message identification field of the IP
frame is used to identify duplicate messages, which may be
discarded. Multipath routing may also improve network reli-
26
Unit 1: Wireless Network Technology
ability by providing redundant message paths. Both possibilities must be considered for industrial wireless networks.
Another problem is the increased latency caused by routing.
Some messages must be delivered to their destination while
the data is still “fresh.” Routing may introduce random delays
that can make data stale. Network configuration must then be
adjusted to avoid routing delays.
Unit 2:
Wireless Network
Standards
The IEEE 802 standards committee has been charged with
developing both wired and wireless data communications network standards. While its work automatically becomes standards in the United States through the American National
Standards Institute (ANSI), it is also submitted to the International Organization for Standardization (ISO) for adoption as
international standards. After a delay of a few months or years
these standards become the series of ISO/IEC 8802 standards,
which have numbers similar to the ANSI standards.
In the definitions or scope of the IEEE 802 committee, the wireless networks are defined by the nominal network transmission distances as described in Table 1.
Table 1. Scope of IEEE 802 Wireless Subcommittees
Subcommittee
Maximum
Distances
WMAN (wireless
metropolitan area
network)
IEEE 802.16
Kilometers
WiMAX®
WLAN (wireless
local area network)
IEEE 802.11
Hundreds
of meters
Wi-Fi®
WPAN (wireless
personal area
network)
IEEE 802.15
Tens of
meters
ZigBee®
Bluetooth™
WiMedia™
Name
Technology
Consortia
27
28
2.1
Unit 2: Wireless Network Standards
Wireless Local Area Networks (WLAN)
Wireless LANs are used to connect computing devices within a
relatively small area. The responsibility for the standardization
of LANs has traditionally been given to the IEEE 802 subcommittees. The standardization of wireless LANs has been
assigned to the IEEE 802.11 subcommittee, and of personal area
networks (PANs) to the 802.15 subcommittee.
Initially, IEEE 802.11 approved a standard that used three different and incompatible technologies: FHSS (frequency hopping spread spectrum), DSSS (direct sequence spread
spectrum), and infrared signaling. Both FHSS and DSSS were
limited to a maximum data rate of 2.0 Mbps. Though implemented by a few suppliers, they were generally unsuccessful in
the market. The 802.11 infrared standard was quite different
from the widely implemented Infrared Data Association
(IrDA) infrared standard and was not commercially implemented at all.
2.1.1
Wi-Fi a/b/g
One of the most successful wireless applications has been the
WLAN, which is enabled by the approval of the IEEE WLAN
standards: 802.11a, b, and g. The extremely large and competitive markets for WLAN in homes and offices has led to high
volumes, with cost reductions leading to selling price reductions. This has turned Wi-Fi into a commodity market. The
IEEE 802.11a and g standards introduced a new spread spectrum technology called OFDM (orthogonal frequency division
multiplexing) that divides the 802.11 frequency band into
many individual sub-channel carrier frequencies, each of
which transmits part of the data. Table 2 describes each of these
network standards.
Unit 2: Wireless Network Standards
29
Table 2. WLAN Comparison
Standard
Designation
Operational
Frequency
Technology
Maximum
Data Rate
Maximum
Distance
IEEE 802.11b
2.4 GHz
DSSS
11 Mbps
100m
IEEE 802.11g
2.4 GHz
OFDM
54 Mbps
100m
IEEE 802.11a
5.4 GHz
OFDM
54 Mbps
100m
IEEE 802.11n
2.4 GHz and
5.4 GHz
OFDM
54 to 400
Mbps
100 to
200m
The low cost of WLANs has made them enormously popular,
especially for the home and small office market where the high
cost of cables and inconvenience of hiding them makes a wireless solution highly desirable. In the larger office, however,
wired LANs still remain more popular because they offer a
higher degree of security and it is easy to install wiring
through raised ceilings and open office partitions. Conventional office wiring for both telephone and LAN is typically
installed simultaneously at very reasonable cost.
Conventional LAN wisdom is already being challenged by
new telephone technologies that place both the LAN connection and the voice connection on the same Ethernet network
using VoIP (Voice over Internet Protocol) technology. With
VoIP, the voice is converted directly at the telephone, or via a
converter for an analog voice network, into an IP data stream
and routed over the IP network. New office installations are
already evaluating and often selecting complete wireless connections for both voice and data.
The release of IEEE 802.11b, which offered a top speed of 11
Mbps, met with sudden and widespread acceptance. With volume purchases of Wi-Fi equipment, prices soon dropped to
commodity levels, and rogue units began to appear in offices.
Very soon afterward, IEEE 802.11a-compatible interfaces also
began to appear, but their incompatibility with 802.11b has
inhibited the market for this higher-speed technology.
30
Unit 2: Wireless Network Standards
Once IEEE 802.11g was approved, there was an immediate
flood of products available to achieve the greater connection
speeds this technology promised. Since Wireless-G operates at
the same frequency as Wireless-B and all of the chips implementing “G” also implement the “B” protocol, this technology’s adoption has been rapid and painless. By early 2005,
Wireless-G had replaced Wireless-B in the consumer marketplace.
Users have become aware that the 2.4 GHz ISM (industrial, scientific, and medical) band may become saturated as a result of
the broad use of Wireless-B and G. This has prompted renewed
interest in Wireless-A, even though its operation in the 5.4 GHz
band makes it incompatible with both B and G. The most likely
scenario is that most silicon implementing Wi-Fi will actually
support both DSSS used for Wireless-B and OFDM as used for
Wireless-A, -G and -N. These will use dual-band radios capable of both 2.4 and 5.4 GHz operation with very little incremental cost. Currently, network equipment for Wireless-A, B, and
G is commercially available at the old introductory prices of
previous WLAN equipment, and Wireless-N in its pre-standard form is available at slightly premium prices.
It is unlikely that WLAN technology will stop at 54 Mbps.
There are proprietary extensions to both G and A that at least
double the data rate, as well as further standardization activity
in the IEEE 802.11 task group n, to produce a standard in the
100-400 Mbps range. Recent actions by the FCC have tripled the
number of channels available to Wireless-A and N in the 5 GHz
frequency band. The market will tend to favor technologies that
have higher data rates as long as the price premium is small. As
the manufacturing technology for this market matures, these
higher-data-rate technologies will advance the natural evolution of these emerging technologies, which has been enabled by
the ever-shrinking size of silicon chips. Eventually, there will
be purchase price parity between wired and wireless network
connections, and wired networks will suffer because of the
high cost of physical cabling. The future is wireless.
Unit 2: Wireless Network Standards
2.2
31
Wireless Personal Area Networks (WPAN)
The IEEE 802.15 subcommittee has been charged with standardizing the emerging technologies of wireless personal area
networks (WPAN). These include Bluetooth, ZigBee, and
WiMedia (IEEE 802.15.3).
2.2.1
Bluetooth
The Bluetooth Special Interest Group originally created Bluetooth; initially as a wireless interconnect method for eliminating wires for use with cellular telephones. While the first
intended application was to replace wiring connecting earphones and microphones, Bluetooth was also intended to
allow cellular telephones to be connected wirelessly to computers when using a cell phone network to connect to the Internet.
The IEEE 802.15.1 subcommittee has adopted the physical and
data link layers of Bluetooth, and organized these layers into
suitable standards format. The upper layers of Bluetooth, however, remain the domain of the Bluetooth-SIG. In these upper
layers lie the primary applications for Bluetooth, including
voice and spontaneous networks. Bluetooth calls its implementation of a limited mesh network a scatternet. In contrast, a piconet consists of a single master and up to seven active slaves.
However, a slave in one piconet may also be a master of a different piconet. A scatternet is a mesh of piconets.
The basic protocol for Bluetooth/802.15.1 is FHSS using a basic
frequency-hopping rate of 1600 hops per second. The number
of frequency channels depends on the country’s frequency allocation in the 2.4 GHz ISM band—seventy-nine channels in
most of the world, but only twenty-three channels in France,
Japan, and Spain.
Bluetooth is defined for three different power ranges in order
to meet requirements for different topologies and distances.
For use as a simple wire replacement in telephony, Power
Class 3 is defined with a maximum power output of 1.0 mW,
covering a distance up to 3 meters. Power Class 2 is for distances up to 100 meters necessary to cover a conference room,
32
Unit 2: Wireless Network Standards
has a defined maximum power output of 2.5 mW. Longer distances that would be necessary for industrial automation are
defined in Power Class 1, with a maximum power output of
100 mW, and requiring “Power Control.” Power Control is a
protocol that requires transmitter and receiver to reduce power
output consistent with the needs to optimize power consumption and overcome interference.
Bluetooth voice applications are optimized to deliver full
duplex time-synchronous voice packets using a time division
multiplexing method. This provides dedicated time slots for
voice packets to be transferred in each direction. Data can also
be transferred in the remaining time after the dedicated time
slots are used.
The Bluetooth scatternet is designed to be spontaneous in
nature. If a node is enabled to join a network, whenever it
moves into radio range of another Bluetooth device it begins to
negotiate to join into a local piconet. If there is already an active
piconet, the new device may join as a slave. If no piconet exists,
the new station and the other negotiate to become the master of
a new piconet. If a piconet is already full (master plus seven
slaves), the new station cannot join unless an existing slave
makes a spot available by moving out of range or by becoming
“parked,” which is an agreement to surrender the network
spot temporarily if the user has not been active for some time
period. Parked nodes no longer communicate with the master
but remain synchronized (i.e., remain in the frequency-hopping schedule.)
Nodes that are denied membership in an existing piconet may
form a second piconet either as a master or as a slave by enlisting one of the existing nodes of the first piconet. The two piconets, together with any others, form a scatternet, which, as we
have seen, is a form of mesh network. The piconets in a scatternet are not synchronized with each other. Messages or voice
packets can be carried between stations on the same scatternet
by being relayed by master stations on each piconet. This
action allows only a single path for the data, in contrast to true
mesh networks in which there can be multiple paths for each
Unit 2: Wireless Network Standards
33
message. Routing algorithms in each master control routing
within the scatternet.
The Bluetooth specification contains profiles for several applications. The purpose of the profile is to define the application’s
protocol. These application layer protocols are in turn supported by the lower-layer data link layer protocols. To understand Bluetooth, it is necessary to understand these
applications, for which it was created. While the underlying
body of the Bluetooth protocol may change, these well-conceived application layer protocols are unlikely to change dramatically for the intended applications. It is well within the
objectives of the Bluetooth SIG to continue supporting these
application profiles even if the underlying data communications changes since the Bluetooth networks are adapted to
coexist with all of the other networks that share the same 2.4
GHz ISM frequencies.
Table 3 describes the specified Bluetooth Application profiles.
Each profile has its protocol specified in the Bluetooth specification. However they are not specified in the IEEE 802.15.1 standard, which only applies to the physical and data link layers.
Generic Access Profile
The Generic Access profile provides the procedures for two
devices to discover and connect to each other when neither of
the two devices has a link established. The profile also provides discovery and connection procedures when at least one
device has established a link to a third device before starting
the procedure. The Bluetooth user will be able to connect a
Bluetooth device to any other Bluetooth device. Even if the two
connected devices don’t share any common applications, it is
possible for the user to use basic Bluetooth capabilities to determine this. When the two devices that share the same application are from different manufacturers, they can still connect
them even if manufacturers call basic Bluetooth capabilities by
different names on the user interface level or implement basic
procedures to be executed in different orders.
34
Unit 2: Wireless Network Standards
Table 3. Bluetooth Application Profiles
Profile
Description
Generic Access
Describes how two Bluetooth stations begin
communicating
Service Discovery
A standardized procedure to locate and identify
Bluetooth services
Cordless Telephony
Services for cordless telephones
Intercom
Services for intercom and paging or walkie-talkie
usage
Serial Port
Services to emulate a serial port connection
Headset
Services to support headphones and a microphone
for full duplex voice communications
Dial-Up Networking
Services to allow a computer to use a cellular
phone or modem as a wireless modem for
connecting to a dial-up Internet access server or for
using other dial-up services, including receiving
data calls
Fax
Services to allow devices to send or receive fax
messages
LAN Access
Services to allow devices to become network nodes
on a LAN
Generic Object
Exchange
Services for transporting data
Object Push
Services for sending, pulling, and exchanging data
File Transfer
Services for browsing and transferring files
Synchronization
Services to support file update processes between
devices
Unit 2: Wireless Network Standards
35
The General Access profile states the requirements for the
names, values, and coding schemes used for names of parameters and procedures on the user interface level. It defines
modes of operation that are not service- or profile-specific, but
are rather generic and can be used by other profiles referring to
this one as well as by devices that are implementing multiple
profiles. The General Access profile defines the general procedures that can be used for discovering the identities, names,
and basic capabilities of other Bluetooth devices that are in a
mode in which they can be discovered. Only procedures in
which no channel or connection establishment is used are specified by the profile. This profile defines the general procedure
for creating bonds (i.e., dedicated exchanges of link keys)
between Bluetooth devices. It describes the general procedures
that can be used to establish connections to other Bluetooth
devices that are in a mode that allows them to accept connections and service requests.
Service Discovery Profile
The service discovery user application in a local device interfaces with the Bluetooth Service Discovery profile client to
send service inquiries and to receive service inquiry responses
from remote devices. Service discovery is tightly related to the
process of discovering devices, and discovering devices is
tightly related to performing inquiries and pages.
Before any two Bluetooth-equipped devices can communicate
with each other the following two conditions are necessary:
•
The devices need to be powered on and initialized. For
initialization, a PIN may need to be provided to create a
link key so the device can be authorized and the data
encrypted.
•
A Bluetooth link has to be created. This may require
that the other device's address be discovered and the
other device be paged.
It may seem natural to consider one device serving as a Bluetooth master and the other serving as Bluetooth slave, but no
36
Unit 2: Wireless Network Standards
such requirement is imposed on devices participating in the
Service Discovery profile. Service Discovery can be initiated by
either a master or a slave device at any point for which these
devices are members of the same piconet. In addition, a slave
in a piconet can also initiate service discovery in a new piconet,
provided that it notifies the master of the original piconet that
it will be unavailable (possibly by entering the parked mode) for
a given amount of time. The Service Discovery profile does not
require the use of authentication and/or encryption. If any of
the devices involved uses any of these procedures, service discovery will be performed only on the subset of devices that
pass the authentication and encryption security that they may
impose on each other. In other words, any security restrictions
for service discovery transactions are dictated by the security
restrictions already in place (if any) on the Bluetooth link.
Cordless Telephony Profile
The Cordless Telephony profile defines the protocols and procedures that are used by devices implementing a “3-in-1
phone.” The “3-in-1 phone” provides an extra mode of operation to cellular telephones, by using Bluetooth to access fixed
network telephony services via a base station. However, the 3in-1-phone service can also be applied to wireless telephony in
a residential or small office environment. This profile includes
making calls via the base station, making direct intercom calls
between two terminals, and accessing supplementary services
provided by the external telephone network.
The following scenarios are covered by the Cordless Telephony
profile:
1. Connecting to the base station so incoming calls can be
routed to the handset and outgoing calls can be
originated.
2. Making a call from a handset to a user on the local
telephone network.
3. Receiving a call from the local telephone network.
Unit 2: Wireless Network Standards
37
4. Making calls between two handsets via the local
telephony network.
5. Using supplementary telephony services provided by
the external network.
Intercom Profile
The Intercom profile is similar to the Cordless Telephony profile, adding only the feature of allowing direct calls between
two handsets that are not using the local telephony network.
This is generally referred to as the “walkie-talkie” profile.
Serial Port Profile
The Serial Port profile defines the protocols and procedures
that are employed by devices that are using Bluetooth for serial
cable emulation. This profile encompasses the scenario in
which legacy applications are using Bluetooth as a cable
replacement, through a virtual serial port. For the purposes of
mapping the Serial Port profile to the conventional serial port
architecture, both devices can be either a data circuit endpoint
(DCE) or a data terminal endpoint (DTE). The protocol is
designed to be independent of DTE-DCE or DTE-DTE relationships. Any legacy application may be run on either device, by
using the virtual serial port as if a real serial cable was connecting the two devices (with RS232 control signaling).
Headset Profile
The Headset profile defines the protocols and procedures that
are used by devices that are implementing the usage model
called “Ultimate Headset.” The most common examples of
such devices are headsets, personal computers, and cellular
telephones. The headset can be wirelessly connected so they act
as the device’s audio input and output mechanism, providing
full duplex audio. The headset increases the user’s mobility
while maintaining call privacy.
The Headset profile provides control over the volume settings
of both the microphone and the speakers. The microphone is
38
Unit 2: Wireless Network Standards
always monophonic, but the speakers may be either monophonic or stereo.
Dial-Up Networking Profile
The Dial-up Networking profile defines the protocols and procedures that are used by devices such as modems and cellular
phones. A cellular telephone may be used by a computer as a
wireless modem for connecting to a dial-up Internet access
server or for using other dial-up services. A cellular telephone
or modem may be used by a computer to receive data calls.
Fax Profile
The Fax profile allows a Bluetooth cellular telephone or
modem to be used by a computer as a wireless fax modem to
send or receive a fax message. For the purposes of mapping the
Fax profile to the conventional modem system architecture, the
wireless telephone or modem is considered a data circuit endpoint (DCE) and the computer is considered a data terminal
endpoint (DTE).
LAN Access Profile
The LAN Access profile defines LAN access by using point-topoint protocol (PPP.) PPP is a widely deployed means of allowing access to networks, which provides authentication, encryption, data compression, and multiprotocol facilities. PPP has
been chosen as a means of providing LAN access for Bluetooth
devices because of the large installed base of devices equipped
with PPP software. PPP is capable of supporting various networking protocols (e.g. IP, IPX, etc.). This profile does not mandate the use of any particular protocol. However, since IP is
recognized as the most important protocol used in today’s networks, the profile provides additional IP-related information.
This profile does not deal with conferencing, LAN emulation,
ad hoc networking, or any other means of providing LAN
access. This LAN Access profile defines how PPP networking
is supported in the following situations.
Unit 2: Wireless Network Standards
39
1. LAN access for a single Bluetooth device.
2. LAN access for multiple Bluetooth devices.
3. PC to PC (using PPP networking over serial cable
emulation).
Generic Object Exchange Profile
The Generic Object Exchange profile defines the protocols and
procedures that are used by the applications that need object
exchange capabilities. These applications are, for example, synchronization, file transfer, or the object push model. The most
common devices that use these applications are notebook PCs,
PDAs, smart phones, and cellular telephones.
Object Push Profile
The Object Push profile defines the requirements for the protocols and procedures that are used by applications that provide
the Object Push model. This profile makes use of the Generic
Object Exchange profile to define the interoperability requirements for the protocols these applications need. The most common devices using this profile are notebook PCs, PDAs, and
cellular telephones. The scenarios covered by this profile are
the following:
1. Using a Bluetooth device, for example, a mobile phone,
to push an object to the inbox of another Bluetooth
device. The object can be, for example, a business card
or an appointment.
2. Using a Bluetooth device, for example, a mobile phone,
to pull a business card from another Bluetooth device.
3. Using a Bluetooth device, for example, a mobile phone,
to exchange business cards with another Bluetooth
device. Exchange is defined as a push of, say, a business
card followed by a pull of a business card.
40
Unit 2: Wireless Network Standards
File Transfer Profile
The File Transfer Profile (FTP) defines the requirements for the
protocols and procedures that are used by applications that
require file transfers. This profile uses the Generic Object
Exchange profile as a base profile to define the interoperability
requirements for the protocols that are needed by the applications. The most common devices that use this profile are PCs
and PDAs. The scenarios covered by this profile are the following:
1. Using a Bluetooth device (e.g., a notebook PC) to
browse an object store (file system) of another
Bluetooth device. Browsing involves viewing objects
(files and folders) and navigating the folder hierarchy
of another Bluetooth device, for example, one PC
browsing the file system of another PC.
2. A second usage is to transfer objects (files and folders)
between two Bluetooth devices, for example, copying
files from one PC to another PC.
3. A third usage is for a Bluetooth device to manipulate
objects (files and folders) on another Bluetooth device.
This includes deleting objects and creating new folders.
Synchronization Profile
The Synchronization profile defines the requirements for the
protocols and procedures that are used by applications that
require that objects stored on both devices be synchronized.
This profile makes use of the Generic Object Exchange profile
to define the interoperability requirements for the protocols
applications need. The most common devices that requiring
synchronization include notebook PCs, PDAs, and cellular
telephones. The scenarios covered by this profile are the following:
1. A computer using a mobile phone or PDA to exchange
PIM (personal information management) data,
including any necessary log information to ensure that
Unit 2: Wireless Network Standards
41
the data contained within their respective Object Stores
is made identical. The types of PIM data include, for
example, phonebook and calendar items.
2. A mobile phone or PDA using a computer to initiate the
previous scenario (Sync Command Feature).
3. A computer using a mobile phone or PDA to
automatically start synchronization when a mobile
phone or PDA enters the radio frequency proximity of
the computer.
2.2.2
ZigBee
ZigBee is the name of a network architecture given by the ZigBee Alliance, an industry consortium that is focusing on promoting the use of low-power networks for applications such as
home automation, industrial automation, building automation,
and toys. While the integrity of the network must not be compromised, the emphasis is more upon power conservation for
battery or other power-sensitive applications. The ZigBee Alliance has supported the development of IEEE 802.15.4 for its
purposes. Along the way, the former HomeRF Consortium has
been dissolved and many of its former sponsors have moved to
support ZigBee.
As of mid-2007, there are several commercial implementations
of ZigBee. Chipcom and Freescale announced their silicon supporting ZigBee. Both operate only in the 2.4 GHz band. Freescale uses their M68HC08 microcontroller family with their RF
Packet Radio chip. The 802.15.4 and ZigBee protocol stacks are
implemented in software. Several other foundries, including
those of Intel, Texas Instruments, Atmel, and Phillips have also
developed silicon. The primary difficulty has been achieving
the low-power specifications necessary to support the battery
and alternate power sources envisioned by the committee.
Ember Corp. has announced that its EmberNet products are
being produced with Chipcom silicon for wireless sensor networking. Ember’s previous products have used more proprietary radios that operate in the same 915 MHz and 2.4 GHz
42
Unit 2: Wireless Network Standards
ISM bands as ZigBee. The release of its ZigBee products is a
significant event.
Millennial Net has announced sensor networking products
that conform to the ZigBee specifications. Millennial also produces its I-Bean products, low power 915 MHz narrowband
radio, as components to be used by product manufacturers.
One of its recent products uses an “energy harvesting” technology from Ferro, in which ambient vibration is used to power
the communications interface, entirely without the need for
batteries. ZigBee’s low-power consumption makes this configuration possible.
Mattel supported the early development work of IEEE 802.15.4
because of its relevance to Mattel’s radio-controlled toys, Leviton to support its wireless lighting controls, and Eaton/CutlerHammer for its relevance to Eaton’s industrial automation
products. These applications have ensured that the requirements for very low power remain among the developers’ priorities.
IEEE 802.15.4 defines a low-level direct sequence spread spectrum radio interface for a network that is capable of transporting data through areas of high electrical noise and metallic
interference at nominal distances up to 100 meters. In addition
to robust noise rejection, the standard stipulates the use of
mesh networking to overcome direct line-of-sight obstructions
and to provide alternative path routing in cases of temporary
network outages. Mesh networking also provides a convenient
way to expand the coverage distances for ZigBee networks,
since the distance limit only applies to the most distant unit.
The ZigBee protocol provides the necessary mechanism for
removing redundant messages received from alternate paths in
the mesh.
IEEE 802.15.4 only defines the communications radio (physical
layer) and protocol (data link layer) for both star (point-topoint) and peer-to-peer topologies. The ZigBee Alliance has
defined the network layer that specifies star, tree-cluster, and
mesh network topologies. Additionally, ZigBee is defining the
application layer profiles for several applications. The initial
Unit 2: Wireless Network Standards
43
application areas for which profiles will be developed are
industrial automation, home control, and building automation.
Additionally, a profile is being developed for Automated
Meter Reading.
IEEE 802.15.4 Technology
The physical and data link layers for ZigBee are defined by the
approved IEEE 802.15.4 standard. Figure 7 illustrates the scope
of the IEEE 802.15.4 standard and ZigBee.
802.15.4 Architecture
}
Upper Layers
IEEE 802.15.4 LLC
IEEE 802.2
LLC, Type I
IEEE 802.15.4 MAC
IEEE 802.15.4
868/915 MHz
PHY
IEEE 802.15.4
2400 MHz
PHY
ZigBee
}
802.15.4
Figure 7. ZigBee Architecture (Source: IEEE working papers)
Two of the primary goals for 802.15.4 are low cost and low
power, which lead to low complexity and simplicity. Negotiating for data rates increases a protocol’s complexity, so 802.15.4
uses just two different data rates: 250 Kbps for high speed and
20 Kbps for slow-speed, very-low-power applications.
44
Unit 2: Wireless Network Standards
Networks of sensors and actuators used in process control tend
to be scattered, while the sensors and actuators used for factory
automation tend to align with the large machines used. Typically, a star network can be expensive in terms of wiring, but a
star network is very simple and inexpensive for wireless networks for either factory automation or process control. However, when the distance for any one device exceeds the
maximum, or when devices need to communicate with other
local devices, peer-to-peer networking may be more efficient.
The form of peer-to-peer networking that is included in the
IEEE 802.15.4 data link layer is very simple and is provided so
as to enable the formation of clusters for tree topology and for
implementing mesh networking at the ZigBee network layer.
ZigBee supports low latency devices. Some devices produce
very little data, such as a pulse when an event occurs, but they
produce it frequently. Photocells that count products and
tachometers that produce speed data are some examples.
802.15.4 provides guaranteed time slots for these types of
devices in which missing data or a single pulse cannot be
recovered.
The basic protocol of 802.15.4 is Carrier Sense Multiple Access
with Collision Avoidance (CSMA-CA). On very simple networks the non-beacon mode can be used, which allows the
occasional collision and retransmission. On critical networks,
beacon mode is used. Here, the node, acting as a network coordinator, arbitrates network traffic to prevent collisions by
assigning nodes to one of sixteen specific time slots. On larger
tree cluster and mesh networks, some nodes are also assigned
to be network routers, and these nodes assign time slots to prevent collisions. All nodes can then sleep (low-power mode)
whenever they are not scheduled to send or receive during a
slot time. Figure 8 illustrates the time slot mechanism for
802.15.4.
Devices then sleep until they are ready to determine if there are
any messages. They awake in time to examine their time slot
and take any appropriate action if there is a message. If not,
then they can return immediately to sleep. It has been estimated that most nodes in a beacon network will remain asleep
Unit 2: Wireless Network Standards
45
GTS 3
GTS 2
GTS 1
15ms * 2 n
where 0 ≥ n ≥ 14
Network
beacon
Transmitted by network coordinator. Contains network information,
frame structure and notification of pending node messages.
Beacon
extension
period
Space reserved for beacon growth due to pending node messages
Contention
period
Access by any no de using CSMA-CA
Guaranteed
Time Slot
Reserved for nodes requiring guaranteed bandwidth [n = 0].
Figure 8. Frame Structure for IEEE 802.15.4 (Source: IEEE working
papers)
approximately 97.5 percent of the time. Sleep invokes the lowpower state so the microcontroller can save power, so the clutter in the frequency band can be reduced, and to avoid most
sources of interference.
Devices are usually addressed with a short (16-bit) address,
which limits the number of nodes in any one subnet to 255. The
subnet is defined by the stations that are managed by a beacon.
In full addressing mode, the node may be directly addressed
by its full 64-bit long address, such as during network setup.
Each message sent to a node is acknowledged by using a
highly efficient short frame. Acknowledgement guarantees
delivery and is the form of confirmed services that is used in
802.15.4.
Low power consumption is enabled by using a very simple
protocol and by allowing all battery-powered remote nodes to
sleep most of the time. The time-slotted services enable each
node to sleep while it is waiting for its slot.
46
Unit 2: Wireless Network Standards
The standard specifies that 802.15.4 technology should operate
at three different frequency bands to accommodate some of the
frequency assignments in the different countries in which the
standard’s approval will be sought. There are sixteen channels
in the 2.4 GHz ISM band that are applicable everywhere in the
world, ten channels in the 915 MHz ISM band that are applicable only in North America, and one channel in the European
868 MHz band. Figure 9 illustrates the frequency allocation for
IEEE 802.15.4.
868MHz/
915MHz
PHY
2.4 GHz
PHY
Channel 0
Channels 1-10
868.3 MHz
902 MHz
Channels 11-26
2.4 GHz
2 MHz
928 MHz
5 MHz
2.4835 GHz
Figure 9. Operating Frequency Bands (Source: IEEE working papers)
The feature of the 802.15.4 protocol that contributes most to
long battery life is the extremely low duty cycle. Each batterypowered network node is intended to sleep 97.5 percent of the
time. The active router nodes that generate the beacons awaken
on the beacon time schedule, and they are awake more than the
end nodes. The coordinator node for the network tends to be a
powered node that is not subject to reduced energy consumption. Figure 10 illustrates the topology of a ZigBee network.
2.2.3
WiMedia (IEEE 802.15.3)
The objective of IEEE 802.15.3 was to develop alternative highdata-rate radio delivery methods for personal area networks.
This work failed to come to agreement, and has been termi-
Unit 2: Wireless Network Standards
47
ZigBee Coordinator
ZigBee Router
ZigBee End Device
Mesh Link
Star Link
Figure 10. ZigBee Network
nated by IEEE, but the work continues through the WiMedia
Alliance and UWB-Forum. The committee defined UltraWideBand (UWB) as its chosen technology. However, UWB is still
new and its use for networks is in an embryonic stage. The
standards committee provisionally approved the use of multiband orthogonal frequency division multiplexing (MODFM),
which simulates the original pulse-position modulated UWB
signal now called DS-UWB (Direct Sequence – UWB) or sometimes CSS (Chirp Spread Spectrum.)
WiMedia was being created to make possible the local connection of high-speed devices such as streaming video over short
distances. This will make it possible to replace hard wires and
therefore simplify the process of connecting devices for home
entertainment. Like other WPANs, the connection distance is
expected to be from a few millimeters to about 10 meters. The
FCC ruling that approved the use of UWB allows the frequency
range to overlap other assigned frequencies. The premise for
this decision was that the signal at each frequency is so short
and low in power that other radio services would view it as
impulse noise. The FCC ruling has forced WiMedia to operate
in the 3.1-10.6 GHz band and to exclude all signals from the
48
Unit 2: Wireless Network Standards
GPS bands at 1.228 and 1.575 GHz. Figure 11 illustrates the frequency assignments for UWB and the restricted power rating
required.
Narrowband (30kHz)
FCC
Part 15 Limit
Wideband CDMA (5 MHz)
UWB (Several GHz)
Frequency
Figure 11. Frequency Band for UWB Radio (Source: IEEE working
papers)
WiMedia is being designed for high data rates from 54 to over
500 Mbps. While such high data rates may not currently be necessary for industrial automation, the prospect of interconnecting systems without wires is often an appealing one. With
more experience in using UWB at these very low power limits,
manufacturers will make greater efforts to extend distance by
increasing average power output.
The Bluetooth SIG has decided to develop a Wireless USB profile based on the use of WiMedia technology. Eventually, this
may get written into an update of the IEEE 802.15.1 standard.
The IEEE 802.15.4a Task Group has, on the other hand,
approved an alternate physical layer using DS-UWB.
2.3
WMAN, WiMAX (IEEE 802.16a)
WiMAX is a robust, higher-power technology that is used for
wireless broadband. It has been developed for long distances,
up to forty kilometers, and for metropolitan locations where
line-of-sight interference is possible. While WiMAX does not
seem well suited to low-level industrial automation applications, it is appropriate for replacing cable used in “home-run”
or other high-bandwidth long cable runs – those identified as
H2 for many fieldbus applications.
Unit 2: Wireless Network Standards
49
WiMAX is specifically addressing the 2.5 GHz frequency band.
It is expected that WiMAX will be deployed across both
licensed and unlicensed bands, for specific needs. This too is a
developing market, but it is one that is based on several years
of experience with Multichannel Multipoint Distribution Service (MMDS) in the 2.1 to 2.7 GHz licensed frequency band.
WiMAX is based on the IEEE 802.16a standard for stationary
nodes, and for mobile nodes in IEEE 802.16e. The WiMAX technology includes mesh network topology that solves the multipath problems encountered by MMDS before it. It is also based
on the use of orthogonal frequency division multiplexing
(OFDM), the same technology that used in 802.11a, g, and n.
Mesh networking also allows suitable signals to be delivered to
areas in which direct line of sight from the primary transmitter
is obstructed. WiMAX is not a low-power standard and is not
intended to use unlicensed frequency bands.
One of the first projected uses of WiMAX is as the wireless
backhaul network for Wi-Fi access points. Backhaul is a telecommunications term for the network infrastructure that is needed
to deliver data to the primary user networks. In this case,
WiMAX is being groomed to connect Wi-Fi access points to the
host network in an all-wireless network.
Another potential use of WiMAX is to be the “last mile” for
delivering broadband to homes and business. This has the oxymoronic name of wireless cable. In this service, hundreds of
channels of television will be delivered, along with two-way
voice (telephony) and data services.
WiMAX is also being proposed for wireless cell phone use
where it is often referred to, unofficially, as 4G wireless.
2.4
Wireless Telephony
Radio has been used for telephone or voice communications
since the 1920s. World War II witnessed the use of walkie-talkie
phones, which were heavy and had very short battery life. Citizen’s Band (CB) radio became popular in the 1970s. The first
wireless telephone network used a technology called Advanced
50
Unit 2: Wireless Network Standards
Mobile Phone Service (AMPS), which is still implemented in
most North American metropolitan areas. AMPS was generally
known as a “car phone” technology because the radios of the
day used high-power technology, which made them large and
heavy. Today, AMPS is often implemented in conventional “trimode” cell phones that use low-power radio. AMPS towers
typically cover a range up to about forty kilometers.
European countries each installed different analog wireless
networks, but soon realized that they needed a system common for all of Europe. The European Economic Community
(EEC) standardized an early all-digital method called GSM (for
“global system for mobile communications”). GSM uses
TDMA (time division multiple access) as well as FDMA (frequency division multiple access) in its protocol. GSM was originally assigned the frequencies 890-915 MHz for the uplink
(mobile station to base station) and 935-960 MHz for the downlink (base station to mobile station) for mobile networks in
Europe. These frequencies are not available for this purpose in
North America. Eventually, the capacity of the system was
improved by adding more frequencies in the 1800 MHz band.
GSM is also implemented in North America in the 1900 MHz
band. You can read much more about GSM on the official website: http://www.gsmworld.com/index.shtml.
The FDMA part involves the division by frequency of the
(maximum) 25 MHz bandwidth into 124 carrier frequencies
spaced 200 kHz apart. One or more carrier frequencies are
assigned to each base station. Each of these carrier frequencies
is then divided in time, using a TDMA scheme. The complex
TDMA and FDMA scheme allows many users to simultaneously use the same frequency on a single base station.
North America evolved more slowly toward replacing AMPS
with digital technology. The first offerings were TDMA
schemes that were not compatible with GSM. All of these are
now being converted to GSM, but at the North American 1900
MHz frequency band (AT&T). T-Mobile has developed a pure
GSM service in the North American 1900 frequency band.
Unit 2: Wireless Network Standards
51
Qualcomm developed its DSSS-based CDMA (code division
multiple access) modulation method, which was adopted by
two U.S. wireless companies (Sprint and Verizon). CDMA
offers a clear migration path to higher digital data rates that are
not directly offered by GSM. In 1998, the ITU (International
Telecommunications Union), which is the United Nationssanctioned standards body responsible for telephony and radio
standards, adopted CDMA-2000, a standard based on Qualcomm’s patents. Its objective was the eventual union of wireless telephony for the world in the 2.4 GHz frequency band.
This plan was called 3G (third-generation) wireless.
CDMA-2000 was doomed from the start since it did not consider all of the varied frequency assignments around the
world, nor did it consider the prior investments in GSM and
TDMA technology. Moreover, the popularity of Wi-Fi in the 2.4
GHz frequency band would suffer if CDMA-2000 were permitted in this same band. However, all is not lost! The revision of
CDMA-2000 is called WCDMA (for “Wideband CDMA”) and
calls for the recognition of regional frequency assignments. It
also calls for the stepwise migration of data rates through some
intermediate steps called 2.5G, since GSM, TDMA, and early
CDMA were considered 2G, or second generation. Specifically,
WCDMA takes into account the use of multiple radio channels
and allows GSM-like TDMA when necessary to maximize the
use of these channels.
The 2.5G steps such as General Packet Radio Service (GPRS)
and EDGE (Enhanced Data rates for Global Evolution) for GSM
and CDMA2000 1xRTT (radio transmission technology) are
interesting. However, they are transitional steps to 3G that will
soon become obsolete. Only WCDMA and the full CDMA2000
1xEVDO (EVolutionary Data Only) specifications will produce
real 3G data rates in excess of 2.0 Mbps. These are the current
standards toward which the cellular telephone industry is
migrating. However, since radio technology is still embryonic,
we can expect more rapid change toward higher data rates,
even before the market settles on a current definition of 3G.
Figure 12 illustrates the evolutionary path of the transition
from 2G to 3G telephone standards.
52
Unit 2: Wireless Network Standards
Figure 12. Evolution of Digital Cellular Technologies
2.5
Convergence of Voice and Data Networks
In case you had not noticed, the evolutions of modern voice
networks and data networks have a great deal in common. At
one time, when voice was carried as an analog signal, the
switched voice network was unique. Gradually, the analog signal was replaced by a digitized stream that was used only in
the long distance networks. Then, the local switching office
was replaced by a digital switch that required the voice to be
digitized as soon as it entered the telephone central office. The
latest telephone technology uses VoIP (Voice over Internet Protocol), in which voice is digitized at the local telephone handset, if there is one, or by a VoIP modem. Wireless telephones
using digital PCS and GSM have always been all-digital. The
commonality now is the use of one universal protocol, IP, for
the voice network as well as for the data network.
The convergence of voice and data networks is expected to
have a profound influence on the development of future network technologies. Voice is the largest single market for any
Unit 2: Wireless Network Standards
53
digital technology, and is responsible for reducing costs for
network elements to very low levels. The voice network has
long been used in cumbersome ways to transport data, as with
telephone modems. With modern VoIP technology, voice and
data naturally mix on the same network, reducing the cost of
implementing the data network by sharing the cost with the
voice network.
Today, controversy swirls over the wisdom of spreading Wi-Fi
“hotspots” for data access with the alternative of making data
connections widely available through the use of 3G technology. The only thing fueling this controversy is the reluctance of
the wireless telephone carriers to make the required infrastructure investment for the propagation of 3G networks and attractively pricing the service. The problem is not technical! If the
wireless telephone carriers understood the rich rewards of
making broadband data service widely available at reasonable
prices, then there would be no need for Wi-Fi hotspots. However, the telephone industry has historically not understood
the need to develop a data utility market, and has priced itself
out of this market on many occasions.
As a result, the cost of 3G is not presently competitive with WiFi solutions. However, the evolution to full WCDMA and
CDMA2000 1xEVDO will occur before the end of 2008, according to the current plans of the wireless telephone industry. By
that time, I expect low-cost versions of 3G to become available
for local converged voice/data networks, much like PBX is
used today. It is this availability that may offer 3G for practical
use in industrial automation. Section 4.5 contains a more technical discussion of this issue.
Unit 3:
Industrial Automation
Requirements
Industrial automation is a difficult market in which to introduce any new technology. The rate of acceptance of any new
technology in industry is relatively slow compared to the commercial, office, or home markets. Even when technology is well
accepted by one of the more visible markets, there are special
environmental problems that must be overcome. Security and
privacy are different in the industrial automation market as
well. Reliability, however, is often the factor where some measurable difference in implementation is required from other
markets.
3.1
Environmental
There are two generally accepted submarkets within industrial
automation:
•
factory automation
•
process automation
Factory automation generally encompasses both machine
shops, where metal cutting is involved in the manufacture of
products; and assembly, where parts are fabricated into finished products. Additionally, in factory automation, materials
handling, movement, or conveying is normally required to
move raw materials, work-in-progress, and final products
within the shop floor and to/from shipping locations. The factory is often dirty, dusty, oily, noisy, filled with vibration, and
electrically noisy because of all the electrical motors used to
power equipment. Temperature, while often uncontrolled, is
usually suitable for human inhabitation. Except for wash-down
conditions, the factory floor is rarely wet.
55
56
Unit 3: Industrial Automation Requirements
In contrast, process automation often occurs in plants located
out-of-doors, and the production is usually hidden from the
plant operators since it usually lies inside pipes, tubes, and
pressure vessels. The products are often fluids, thus leading to
the label fluid process industries. Often the products and the
intermediates are volatile and sometimes flammable or poisonous to humans. In most cases, the fluids are corrosive as well.
Here too, electrical noise is usually high because of all the electrical motors used to power mixers and pumps. Temperature is
uncontrolled except for the places, typically called control
rooms, which are intended for human habitation. The process
area is typically subject to all types of atmospheric conditions,
including rain, snow, ice, wind, and direct sunshine.
3.2
Security
The security of an industrial automation network means the
network is protected against espionage, sabotage, or attack.
This implies a freedom from risk or danger from outside
sources. It also means the network may be depended upon to
continuously do its job of delivering data when and where it
was intended. Where process or human safety is one of the network tasks, that safety is ensured.
The term security is often used to include both privacy and reliability, but this dual usage is incorrect. We discuss privacy and
reliability in the next two sections. However, many of the solutions for making a network more secure will also have the side
effect of increasing privacy as well.
3.3
Privacy
Privacy is defined as the quality or condition of being secluded
from the presence or view of others. It might also be described
as the state of being free from unsanctioned intrusion, or being
concealed. If network security eliminates unauthorized intrusion, then it has effectively provided a privacy solution as well.
Even when the network is adequately protected against external intrusion, it must still be protected against persons or sys-
Unit 3: Industrial Automation Requirements
57
tems accessing or changing data to which they are not
authorized. Often, the information technology (IT) staff in a
particular business area will have open access to the business
network, but they could do unintended damage if their access
were extended to the automation network. For example, a common IT procedure is to notify all users of some company-wide
event such as a server being rebooted. In the Windows operating system, such notices are sent using the NET SEND command. This creates a broadcast message on the network that
passes through all network switches. A router that has been
enabled to prohibit broadcast messages through stateful
inspection will block such messages. Stateful inspection is a
firewall that keeps track of the state of network connections
(such as TCP streams) travelling across it.
Automation network access by authorized persons and programs must be allowed, but not by unauthorized persons or
programs, even though they may have open access to the entire
business database and network. Blocking network access is
generally assigned to devices called firewalls. For example,
firewalls are conventionally used to isolate a business’s network from the Internet. Firewalls work by blocking access to IP
port addresses. Additionally, they may authenticate users by
requiring a log-on.
3.4
Reliability
Reliability means dependability. Network failure often means
the failure of the automation system upon which control of the
process or machine depends. When a network fails some type
of system response is usually required to bring the process or
machine to a safe state, depending on the nature of the data
being passed on the network. For example, if all control occurs
only in field devices and they do not depend on the network,
then network failure need not cause control to cease, but it
should produce an alarm to notify human operators. If the process cannot run when the network fails, then some fail-safe
mechanism must be implemented to bring the process or
machine to a safe state.
58
Unit 3: Industrial Automation Requirements
Higher reliability in cabled networks can often be achieved by
using premium components such as better cable, connectors,
and terminations as well as network electronics of higher quality or greater ruggedness. The only equivalent reliability
achievable in wireless networks is to use higher-quality or ruggedized network electronics. Higher power radio can often be
used to overcome interference, but most of the time, radiated
power is limited by governmental regulations.
Fault-tolerance is a solution for improving network reliability.
Networks that are fault-tolerant provide more than one network path between any two nodes. Mesh networks are specifically a fault-tolerant solution for wireless networks.
3.5
Power
Industrial wired networks are generally expected to deliver
power to each node, as well as to carry the network signals. In
many process plants, the network is also expected to be intrinsically safe, meaning that a cable break will not cause flammable gases to ignite. Wireless networks definitely have the
advantage of not using wire and are inherently safe, but powering wireless nodes remains a problem.
Battery power is not a well-accepted power source for primary
control devices, except to provide backup power. This attitude
may eventually change, but a better primary power source
must be provided for wireless devices. Local AC, or sometimes
DC, power is always an option, and wiring for power costs
much less than wiring for communications, since local power
sources are usually available.
Unit 4:
Application of
Wireless Networks to
Industrial Automation
4.1
Politics of Wireless
Most of us in industrial automation would like solutions to be
based upon sound engineering principles and open, competitive procurement practices. In other words, we would like to
purchase our solutions/applications from the supplier that has
the best product at the right price and delivery for our needs
and supplies the type of maintenance services we require.
Unfortunately, it just doesn’t work that way! All too often, suppliers have a vested interest in milking the last quarter-year of
product life from existing products, or most often perceive the
need to make any new product backwards-compatible with
existing products to maintain their customer base. To this end,
they participate in standardization activities and industry consortiums with the stated mission of embedding their technology into the resulting standards and documents. This is not
evil; it is business.
Wireless is a disruptive technology, as Peter Drucker described
the term in his book: Innovation and Entrepreneurship: Practice
and Principles (Harper & Row, 1985). According to Drucker,
business is conducted differently after a disruptive technology
is introduced. Leading suppliers to an industry will always
resist disruptive technologies since they tend to render existing
products obsolete before the end of their desired life cycles.
Disruptive technologies offer new opportunities for new or
secondary competitors within an industry to assume leadership by displacing entrenched suppliers. In process control, the
introduction of the distributed control system (DCS) was such
59
60
Unit 4: Application of Wireless Networks to Industrial Automation
a disruptive technology, and Honeywell emerged as the industry leader, displacing previous leader Foxboro. Fieldbus is
another example. Emerson has successfully played its Fieldbus
leadership card and has emerged as the process control DCS
leader, displacing Honeywell and pushing Foxboro further
down the stack. Currently, no leading process control supplier
has fully embraced wireless, which indicates that we are not
yet on the brink of a disruptive revolution. However, both
Honeywell and Emerson have now introduced their second
generation wireless networks and devices with a promise to
upgrade them to the future ISA100 wireless standard when it
becomes available.
No factory automation supplier has yet embraced wireless
either, although Eaton has been an early sponsor of the development of IEEE 802.15.4/ZigBee. The transition of industrial
networks to an Ethernet base in this industry is now in its market development phase. Therefore, interest in wireless is currently at a low ebb. However, even before Ethernet solutions
become the de facto standards in factory automation, wireless
will have a dramatic effect.
The politics of wireless tend to work against the conservative
suppliers in the industrial automation industry, both in process
control and factory automation. Wireless has already invaded
the home: more than 65 percent of home networks are wireless.
Office networks are also rapidly expanding with wireless links,
especially since the equipment is very low-cost, highly reliable,
and easy to configure. Wireless is spreading much as the personal computer did in the PC revolution of the mid-1980s. If
this parallel holds, then wireless will eventually invade the
shop floor in manufacturing – with or without the product support of current suppliers. As the major automation suppliers
resisted the PC, the HMI (human-machine interface) market
rose to meet the industry’s needs. Something similar can happen with wireless.
To give order to wireless, there are standards. Unfortunately,
there are too many standards, a problem not dissimilar to the
fieldbus situation. In addition to the declared standards work,
there are also separate industry consortiums and vendors’ pro-
Unit 4: Application of Wireless Networks to Industrial Automation
61
prietary products. These are also part of the politics of wireless.
At the moment, no single wireless solution exists that is capable of solving all connection problems in all of industrial automation. The remaining part of this chapter reviews the
potential of each of these solutions to play some part in the
wireless future for the industrial automation market.
4.2
Wi-Fi
Wi-Fi can be used anywhere that wired Ethernet can be used,
and in many other locations too hostile, expensive, inconvenient, or cluttered for physical wiring. However, Wi-Fi does
have many limitations that may make it the non-optimal choice
for a wireless industrial network. Wi-Fi, like Ethernet, is
designed for applications that are permanently connected or on
all of the time. Wi-Fi is not meant for mobile applications,
although recent modifications to IEEE 802.1d allow some
mobility between the switches of a single wired network. There
are also some recent implementations of wireless switching for
Wi-Fi networks. However, for a device such as an automated
guided vehicle or forklift truck, which is in constant movement, these solutions require excessive time and network overhead.
Wi-Fi is also not designed for low-power applications. Notebook computers using Wi-Fi on battery operation typically
experience a rapid power drain that often reduces their effective battery life by 50 percent or more. While Wi-Fi does not
require high transmit power, the protocol for the network
expects to find the Wi-Fi radio ON all of the time. Field instruments or other measurement devices would need a local source
of power to effectively use Wi-Fi.
Recent advances in wireless switching for Wi-Fi access points
effectively turn the access point into a network switch. Traditional Wi-Fi access points are more like Ethernet hubs that
broadcast all messages received from the wired network to all
Wi-Fi devices within their radio range. It is the responsibility of
the wired switch in the network above the access point to use
its IEEE 802.1d spanning tree bridge protocol to filter out all mes-
62
Unit 4: Application of Wireless Networks to Industrial Automation
sages not intended for stations connected through that access
point. Access points are now available to perform these switching functions. However, mobile devices must be able to initiate
a message whenever they move from one location to another in
order for the network spanning tree to remove them from the
old location and establish them in the new location.
Wi-Fi networks are intended for high-bandwidth usage, varying from a low of 1.0 Mbps to today’s limit of 54 Mbps, and
perhaps soon to 480 Mbps or more, given the future IEEE
802.11n standard. High bandwidth usually requires more
power than lower bandwidth, which makes Wi-Fi less appropriate for battery or other low-power applications. Obviously,
Wi-Fi is used on battery-operated notebook computers, but
here the Wi-Fi component uses far less energy than the microprocessor, and is therefore not the central issue in battery life.
Of lesser importance is the fact that the Wi-Fi spectrum is quite
crowded with wireless LAN traffic, cordless telephony, Bluetooth, microwave ovens, and other unlicensed uses. Wi-Fi has a
range of about 100 m or less depending upon the topology,
which is typical of most networks operating in the 2.4 MHz
ISM band.
The fact that Wi-Fi can easily be substituted for Ethernet in
almost any application means it will be able to be used without
modification in industrial automation applications where
Ethernet is already accepted, such as to replace Modbus/TCP,
FOUNDATION™ Fieldbus HSE, EtherNet/IP, or PROFINET.
Obviously, there will be performance, privacy, and security
issues that these network designers did not consider when
high speed wired Ethernet was to be the base network, but
many early applications will ignore the obvious problems.
Eventually, the network sponsors will need to address performance, privacy, and security issues for Ethernet-based networks using Wi-Fi.
4.3
Bluetooth
When Bluetooth was first announced, many thought that it
would naturally become the most favored wireless network for
Unit 4: Application of Wireless Networks to Industrial Automation
63
industrial automation. However, part of that premise was that
the cost of the Bluetooth node would quickly drop to only a
“few dollars” as the volume of Bluetooth implementations
exponentially increased. The key to high volume in Bluetooth
sales was its core application: a wireless headset for cell
phones. Unfortunately, the cost of wireless telephone headsets
still lingers in the $40-80 price range, dramatically limiting
Bluetooth’s sales volume. The commodity market for Bluetooth
has yet to emerge, and the cost of the chip remains high relative to other wireless choices.
A closer look at the full Bluetooth suite has also revealed significant overhead posed by its software, which has again complicated Bluetooth’s acceptance for uses other than the original
purposes for which it was invented, and for which profiles are
already available. While the overhead does decrease efficiency,
the more important factor is the large memory requirements
for embedded applications. This has enabled “improved” protocols to be developed for other applications such as ZigBee for
industrial automation and WiMedia for streaming video. The
net effect is to reduce the market for Bluetooth silicon, keeping
its selling price high.
The future for Bluetooth lies in the hands of the Bluetooth SIG,
the organization supporting the development of the Bluetooth
standard. There is considerable interest within Bluetooth SIG
in simplifying the Bluetooth stack for applications that do not
require voice or streaming video. Suppliers that are already
committed to Bluetooth for industrial automation and other
similar applications drive some of this interest. Other members
of the Bluetooth SIG want to increase the performance of Bluetooth for more traditional data uses such as wireless USB and
for larger piconets. Read more about Bluetooth at the Bluetooth
websites: http://www.Bluetooth.org and http://www.Bluetooth.com.
Notice that Bluetooth does not have a profile for industrial
automation applications. This is mostly because the Bluetooth
SIG was focused on telephony-like applications. For Bluetooth
to become useful for industrial automation, suitable profiles
would need to be developed that are specific to this application.
64
4.4
Unit 4: Application of Wireless Networks to Industrial Automation
ZigBee
ZigBee was designed for industrial automation as one of its
core markets. The protocol was designed for significant sleep
time in excess of 97 percent, enabling long battery life. Intermittent use of the spectrum also reduces the opportunity for interference. The moderate data rate and bandwidth were designed
for industrial automation applications to keep energy consumption low during transmission. However, ZigBee implements only DSSS on a single channel in the 2.4 GHz domain,
and does not offer frequency hopping that has been successfully used for early industrial automation products.
The purpose of this book is not to sell any particular technology—that is the task of commercial equipment suppliers. However, it does appear that ZigBee fulfills many of the market and
technical requirements for many of the industrial automation
applications where wireless data transmission to/from field
devices is involved. At this point, acceptance of ZigBee is quite
uncertain. One of the supporting founders of the ZigBee Alliance is Leviton, one of the world’s largest makers of AC receptacles and switches. However, they have not yet announced
volume products for ZigBee that could elevate chip production
to commodity levels.
Both Honeywell and Invensys are supporting the ZigBee Alliance, but more for their respective process control or building
automation products. Eaton/Cutler-Hammer also supports
ZigBee, but it also has not announced any product intentions.
Certainly, ZigBee does not support the high-data-rate backbone networking functions. Therefore, any entirely wireless
industrial automation solution would need to support ZigBee
and some other wireless broadband protocol.
There is still some controversy over the ability of ZigBee’s
direct sequence spread spectrum protocol to adequately reject
industrial electromagnetic noise, vs. use of frequency hopping
spread spectrum as used by both Honeywell and Accutech in
their commercial process control field instruments. While both
DSSS and FHSS reject noise, at least these two suppliers advocate that FHSS is more robust.
Unit 4: Application of Wireless Networks to Industrial Automation
65
Furthermore, ZigBee alone is not enough, since it only defines
the basic means of data communications. More complete upper
layers such as for FOUNDATION™ Fieldbus, Profibus,
DeviceNet, or LonWorks are necessary before a technology
such as ZigBee can be used in an industrial automation system.
You can read more about ZigBee at the ZigBee Alliance website
(http://www.zigbee.org) and also at the IEEE 802.15.4 website
(http://www.ieee802.org/15/pub/TG4.html).
4.5
ISA100 Standard for Wireless Industrial
Networks
ISA Standard and Practices committee 100 (SP100) was chartered in 2005 to prepare a standard for wireless industrial networks. It appears that the first release, designated as
ISA100.11a and targeted to the needs of the process industries,
will be completed in 2008. Most of the major automation suppliers are active contributors to this committee. The categories
of wireless applications have been classified by the ISA SP100
committee as shown in Table 4.
Table 4. Categories of Wireless Applications
ISA100 is targeted to be a family of compatible networks for
the industrial manufacturing environment. The first release,
ISA100.11a is being developed to be the universal wireless network to support FOUNDATION™ Fieldbus, HART, and Profibus-
66
Unit 4: Application of Wireless Networks to Industrial Automation
PA when accessing these devices via a wireless interface. It is
expected that their respective consortiums will create wireless
applications to work with the ISA100.11a base network. As of
mid-2007, the Fieldbus Foundation has agreed to base its wireless version of FOUNDATION™ Fieldbus on ISA100; HART
Communications Foundation has decided to release its own
unique wireless interface, but is working with ISA100 to seek
some type of interoperability. PROFIBUS Nutzerorganisation
has agreed to support WirelessHART, but is actively contributing to ISA100 with the expectation of basing wireless Profibus
on ISA100 technology.
ISA100.11a is being designed to support non-critical applications in the process industries, but is not exclusive in this support. This first release supports low-speed closed loop control
with cycle times no faster than one second, and latencies no
smaller than 100 ms. Future releases will be targeted for higher
speed control loops and lower latencies suitable for discrete
controls in factory automation.
Not only are ISA100.11a compliant field instruments expected
to be on the market by the end of 2008, but modules also will be
available to attach to existing wired HART instrument 4-20mA
cabling to access the digital HART data available on this wiring. All HART data is obtained by specification of the HART
DD (Device Description,) a subset of EDDL. The modules will
be able to draw power from the 4-20mA wiring just like the
HART instrument. The HART data can then be used on an
ISA100.11a compatible handheld device, or may be routed
through the ISA100.11a network to a gateway device connected to a host system.
4.5.1
ISA100.11a Technology
Like ZigBee and WirelessHART, ISA100.11a is based on the
use of IEEE 802.15.4 chips operating in the 2.4 GHz frequency
band at very low duty cycles to enable long life in battery-powered nodes. ISA100.11a additionally specifies channel hopping,
or fixed frequency changing among the 16 channels defined for
IEEE 802.15.4. One of these channels is not available in all
Unit 4: Application of Wireless Networks to Industrial Automation
67
countries, and may be excluded from use for any particular
installation. Nodes may be configured to be leaf-nodes with no
mesh routing to maximize battery life, or full mesh nodes that
are capable of routing messages toward the gateway. Channel
hopping provides frequency diversity to avoid interfering signals, and mesh routing provides spatial diversity to eliminate
fade or multipath effects. Mesh routing also provides the distances and topological paths needed to reach all parts of the
plant. Additionally, ISA100.11a uses TDMA (Time Domain
Multiple Access) for longer messages to maximize use of the
media. Routing within the ISA100.11a field network is provided within the upper portion of the Data Link Layer.
In order to provide long battery life for battery-powered nodes,
ISA100.11a nodes are asleep most of the time. Nodes awake on
a variable schedule to transmit, receive, or relay (route) data.
The awake/sleep schedule is a configurable variable to allow
the network to be tuned somewhere between maximizing battery life and minimizing latency time.
Unlike IEEE 802 standards, ISA100.11a is a full stack solution
to the industrial network problem. A lightweight version of
Internet Protocol Version 6 or IPv6 (6LoWPAN, IETF draft
standard RFC4944) is used at the Network Layer, providing IP
access to field devices, if desired. The Network Layer provides
routing compatibility outside the ISA100 network. For reasons
of efficiency, the Network layer defines short addresses (16bits) that are used for normal data exchange among ISA100.11a
devices. Short addresses are also used to make devices inaccessible to normal IP access. However, since the short addresses
are subsets of the actual device 128-bit IPv6 address, it is possible to access field devices via a long IP address when necessary. Use of this type of addressing eliminates the maximum
number of notes on an ISA100 plant network.
The Transport Layer provides acknowledged and unacknowledged transfers and allows retries for data not successfully
delivered to its destination. Unlike TCP, the ISA100 Transport
Layer does not attempt to optimize message routing or other
inefficiencies of TCP. Block data transfer, unicast, and multicast data transfers are supported.
68
Unit 4: Application of Wireless Networks to Industrial Automation
The ISA100.11a Application Layer not only provides all of the
usual read/write and upload/download functions, but also
adds an object-based Upper Application Layer based on access
to field data parameters using EDDL (Electronic Device
Description Language) that is standardized by IEC 61804. Use
of EDDL is designed to make ISA100.11a immediately compatible with FOUNDATION™ Fieldbus, HART, and Profibus, all of
which have EDDL compatible data access on their wired networks. Additionally, OPC has also adopted EDDL for its
object-based data network access.
For networks in which devices may require the use of protocols
that cannot be resolved with the native ISA100 protocols, messages in the native device protocol can be encapsulate using
this protocol option. The encapsulated message may then be
carried on the ISA100 wireless network between the Gateway
to the host or upper level network and the field device.
Security has been interwoven into the design of ISA100.11a
from the beginning. In general, ISA100.11a security uses symmetric AES-128 or 256 encryption keys or asymmetric public/
private encryption keys for data exchange. The use of carefully
designed encryption allows the use of ISA100.11a to be protected from unauthorized access or interference. Additionally,
ISA100.11a is designed to be configured over the air without
requiring a specific configuration device.
ISA100 has been designed for installation in a wide variety of
plants from small to very large. One of the options is to use
field routers to extend the network to dimensions and locations
that may otherwise be unsuitable for wireless networks.
This forward-looking standard also allows using different
chips at the physical layer in the future to obtain the performance that will be enabled as changes in silicon become available. It has been based on available work from other standards
committees including work done by the IETF (Internet Engineering Task Force) as well as IEEE 802 in various committees.
Unit 4: Application of Wireless Networks to Industrial Automation
4.6
69
WirelessHART
Starting in 2006, the HART Communications Foundation
launched its program for HART version 7 that includes WirelessHART protocol. The objectives for WirelessHART were to
provide a wireless link to channel the HART data captured in
more than 5 million field instruments without a digital connection to the control system to which they are attached. Additionally, Wireless HART is to provide the protocol to be used to
build new HART instruments with a wireless interface for use
by field technicians and for connection to host systems.
The primary benefit for WirelessHART is the ability to network
the millions of existing wired HART devices that currently
have no connection to a DCS or any other network. This
requires a Gateway device to gather information from the
WirelessHART network and supply it to a host system. A common interface protocol between the WirelessHART Gateway
and a host network is being defined for host networks running
FOUNDATION™ Fieldbus or Profibus.
WirelessHART devices are expected to be marketed as full
wireless instruments and as adapter modules for connecting to
existing wired HART instruments.
4.6.1
WirelessHART Technology
WirelessHART, like ISA100.11a and ZigBee, also uses the IEEE
802.15.4 chip in the 2.4 GHz band with low duty cycles for long
life with battery-operated nodes. Like ISA100.11a, WirelessHART also uses channel hopping among 15 of the available
channels and mesh routing. In WirelessHART, all nodes are
capable of routing. The WirelessHART protocol has simple
Network, Transport, and Application Layers. WirelessHART
security is implemented using 128-bit AES encryption. Different encryption keys are used for joining the wireless network
and for transferring data.
WirelessHART nodes are asleep most of the time to conserve
battery life for battery-operated nodes. Each WirelessHART
70
Unit 4: Application of Wireless Networks to Industrial Automation
node awakens on a fixed 10ms time schedule to transmit,
receive, or relay data.
The network layer is responsible for routing using graph routing technology. Routing provides redundant paths for reliability, and is optimized for minimal latency.
The WirelessHART Transport layer is defined to allow long
data transfers to be segmented during transmission. Broadcast,
multicast, and unicast transmissions are all supported. Reliable
block data transmissions are supported, including un-acknowledged and acknowledged data transmissions with retries in
case the transmission was unsuccessful.
A formal Application layer is not defined, but HART 7 instruments may transmit data values by request, upon significant
change, or upon crossing a critical threshold value. HART values are addressed using the traditional HART DD’s (Device
Descriptions) or the newer EDDL shared with FOUNDATION™
Fieldbus and Profibus-PA. Additionally, WirelessHART configuration settings are also addressed by new DD’s defined for
that purpose.
The common Gateway interface between WirelessHART and
host networks for FOUNDATION™ Fieldbus and Profibus has
not yet been defined (as of late 2007.) The purpose of this interface is to make HART devices connected through the WirelessHART network accessible to controllers using only the
HART commands that are based on EDDL or its DD subset.
Modules for WirelessHART will be available to attach to existing wired HART instrument 4-20mA cabling to access the data
available on this wiring. All HART data is then available by
specification of HART DD (Device Description). The modules
will be able to draw power from the 4-20mA wiring just like
the HART instrument. The HART data can then be used on a
WirelessHART handheld device, or may be routed through the
WirelessHART network to a gateway device connected to a
host system.
Unit 4: Application of Wireless Networks to Industrial Automation
71
Because the WirelessHART network is limited to only handling
HART data, the security aspects of WirelessHART are much
less complex than those for ISA100. The WirelessHART interface modules and WirelessHART instruments are expected to
be configured on the bench with the instrument name, tag, and
security key. A conventional HART handheld configurator
may be used for all of these operations. This means that WirelessHART instruments must first be configured by attachment
to a 4-20mA connection, even if this connection is not used during operation. WirelessHART instruments cannot be initially
configured over the wireless network.
4.7
Comparison: WirelessHART vs. ISA100.11a
Since WirelessHART and ISA100.11a were written to serve the
same general process control market needs, it would seem that
they should be compatible. However, they were written by two
independent organizations, with different goals and with no
interaction between them until the HART 7 specification was
completed. Although the ISA100.11a development effort was
public and open, the development of WirelessHART was private and closed. While all of the information from the ISA100
development was available to the WirelessHART development, those portions of ISA100 that were not yet completed
were developed by the HART 7 working group to meet only
the needs for process data acquisition and control. The WirelessHART effort focused on a “wireless extension” of the
HART protocol for the process industries, while ISA100
focused on development of an industrial wireless network to
be applied in a consistent and unified manner for both process
and discrete manufacturing industries. Additionally, the
ISA100 committee recognized the need to support multiple
fieldbus protocols for complex intelligent field devices such as
for FOUNDATION™ Fieldbus and Profibus-PA. This is where
and why the divergence happened: in the layers between the
Application Layer and the Data Link Layer.
The WirelessHART and ISA100.11a Physical Layer and the corresponding MAC portion of the Data Link Layer are identical.
Even the way in which the channel hopping is done is almost
72
Unit 4: Application of Wireless Networks to Industrial Automation
identical. ISA100.11a allows all 16 channels to be used in countries where all 16 channels are available, but with the ability to
exclude any channel. WirelessHART specifies only the 15 channels available worldwide. The channel hopping pattern is different, but ISA100.11a allows using the same pattern as
WirelessHART.
WirelessHART uses a fixed 10ms time slot (awake time).
ISA100.11a uses a variable time slot in order to adapt to specific
application requirements, including the same fixed slot time as
WirelessHART. Variable slot time allows “tuning” the network
so it may be optimized for performance and overall scalability.
Similarly, ISA100.11a allows either AES-128 or 256 bit security
keys to be used, while WirelessHART has been simplified to
use only AES-128 bit security keys. Additionally, ISA100
allows the use of session key rotation to improve security,
while WirelessHART does not.
Both WirelessHART and ISA100.11a use 16-bit local addresses
that are the lower 16-bits of a larger unique address. That
larger unique address is 64-bits for WirelessHART and 128-bits
for ISA100.11a, in order to meet the needs to support IPv6 for
the Network Layer. The network address choices show that
WirelessHART extends HART to the wireless domain, but
ISA100 is designed for the future to allow a more universal
technology to accommodate all protocols.
All of the above are minor differences in which ISA100.11a can
be configured to be a “superset” of WirelessHART through the
Data Link and MAC layers. However, the layers above these
are quite different.
ISA100.11a uses a full IETF 1 standard Network Layer for routing messages to/from field devices. No such external routing
exists for WirelessHART. When the Network Layer protocol is
applied, the resulting frame format of messages from each of
the two networks is very different, making them incompatible.
1. Internet Engineering Task Force, the standards body controlling Internet
standards.
Unit 4: Application of Wireless Networks to Industrial Automation
73
Use of the IETF standard at the Network Layer allows ISA100
messages to be encapsulated and routed over any backbone
network based on Internet Protocol. Additionally, use of this
routable Network Layer expands the maximum size of an
ISA100 network to thousands of field devices, while a WirelessHART network is limited to 250 field devices. Managing
addresses for ISA100 networks in large plants will be simpler
than managing addresses for WirelessHART networks in the
same plant.
Is there hope for convergence? Not without a lot of pain and
grief. The work on WirelessHART is excellent work, but narrowly focused on only one application, and responding to the
goal to be completed as soon as possible. If users make the
decision to wait a few months for ISA100.11a compliant
devices to be on the market, and do not make a commitment to
purchase WirelessHART, then economics will settle the issue.
If, on the other hand, WirelessHART is as popular as wired
HART, then there can be no good solution. Inserting the WirelessHART protocol stack into the ISA100 protocol (a dual-stack
approach) should not be regarded as “convergence.” Replacement of the ISA100.11a protocol with WirelessHART protocol
should be viewed as a capability and security reduction, not
convergence.
Solution to this “multi-standard” problem is not technical, it’s
commercial. Some suppliers have already made the decision to
offer WirelessHART compliant instruments and WirelessHART adapters for wired HART instruments with free
upgrade to ISA100.11a when it becomes available. Users of
these devices will not have a compatibility problem. Other suppliers of WirelessHART instruments and wired HART adapters have not yet made any similar pledge because they have
not received any demands from users for convergence or
ISA100 compliance. This author’s opinion is that all suppliers of
WirelessHART instruments, adapters, and gateways should
offer an ISA100.11a conversion option that should be either
free or at low cost.
74
Unit 4: Application of Wireless Networks to Industrial Automation
4.8
3G/4G for Automation
Why is it appropriate to consider wireless telephony for automation applications? Notice that most of the emphasis on 3G
centers on its potential for high-data-rate digital data transmission. The telephone industry wants to use 3G for video, chat,
graphics, and e-mail applications to supplement voice revenues. Of course, there will be many applications for these services within industry, but 3G and its practical implementation,
WCDMA, share many of the desirable characteristics of an
industrial wireless network as well:
•
Low power consumption
•
High data rate
•
More than adequate distance coverage
•
High-volume silicon production leading to low cost
•
High levels of security protection
•
Confirmed/connected services
•
Low latency
Since 3G is being implemented for battery-powered handheld
cellular PCS (personal communications services) telephones,
and long battery life is important to consumers, service is
designed to conserve energy during active radio connection.
Chips for 3G will draw little more energy than the chips for
Bluetooth or ZigBee. In fact, requirements for long standby
power and talk time led to the same design choices as for these
technologies. Only ZigBee’s scheduled communications cycle
(beacons) can achieve lower duty cycles, and beacons can be
implemented on top of 3G technology. Short Message Service
(SMS), which is now integrated into many cellular telephones,
is an example of such a low-power protocol.
The data rate for 3G has been tested as exceeding 2.4 Mbps
(CDMA2000 1xEVDO) in stationary (non-mobile) applications,
with a requirement for at least 2.0 Mbps. Even in mobile appli-
Unit 4: Application of Wireless Networks to Industrial Automation
75
cations, 384 Kbps must be achieved. These rates are achievable
for distances between the telephone and the cell tower of about
2.5 Km or less, which is much longer than most industrial automation network requirements.
Much has been written about industrial automation needing to
use silicon that has been developed for high-volume markets.
With annual wireless cell phone sales exceeding 460 million
units (according to Gartner’s Mobile Terminal Market Shares:
Worldwide, 2Q03) sufficient volume is unquestioned. Most of
the cost of cell phones is related to the user interfaces: color
LCDs and keyboards. Industrial automation applications need
only the radio and the protocol chip, but not the usual 3G telephone applications.
4G is the designation for “some future technology” that is
expected to replace 3G. This has yet to be defined, but many
expect that WiMAX will serve this need.
Some readers of this book will wonder why GSM and its derivative digital communications standard, EDGE, are not included
in this discussion. EDGE is only an interim 2.5G evolution of
GSM and TDMA telephony to make it possible to achieve a
peak data rate of about 384 Kbps. The longer-term 3G evolution for GSM is WCDMA, which is included in this discussion.
The security of the CDMA and WCDMA cell phone protocol is
naturally high since it is a sequence of packetized data sent
using DSSS. Intercepting this protocol’s signal would require
knowing the exact chipping sequence out of millions of combinations. Assuming that this barrier was broken, data messages
would need to be encrypted using the IEEE 802.1x standard,
which recently adopted the Advanced Encryption Standard
(AES) for its long encryption key.
Industrial automation networks have generally required some
type of confirmed service to validate that critical messages
have been delivered to the proper destination. This is a characteristic of telephony and its connected services. However, for
industrial automation uses, a simpler response protocol such
as in ZigBee might be used.
76
Unit 4: Application of Wireless Networks to Industrial Automation
In command and control applications, a decision to turn on or
off is critical and must not be delayed. Likewise, in telephony,
voice cannot be delayed. 3G is designed to deliver digitized
voice packets that have low latency, which would satisfy
almost all of industrial automation applications.
However, it is not enough to say that 3G is suitable for industrial automation. One of the network standards organizations
now active in industrial automation must specify exactly how
it is to be used. This has not yet been done for 3G any more
than it has for any other wireless standard such as for Bluetooth or ZigBee. Organizations such as the Fieldbus Foundation, ODVA, Modbus Organization, and/or Profibus
International need to add their upper-layer protocols on the
wireless base and test them to confirm their suitability for
industrial automation applications that are at least within their
field of usage. So far, none of these groups is even considering
ZigBee, Bluetooth, or 3G.
Unit 5:
On the Bleeding Edge
Wireless is now the area of networking receiving the greatest
investment in new technology development. There is a great
deal known about the propagation of conventional narrowband and spread-spectrum wireless in the frequencies up
through about 5 GHz. Many promising wireless technologies
have also failed in the marketplace for which they were
intended. All of this is typical for an embryonic market such as
wireless networking. The following technologies are not yet
ready for full commercialization, much less for the industrial
market, which demands field-proven technologies. We often
refer to these not-yet-fully-developed technologies as the
bleeding edge. However, we must always be aware that market development times have dramatically shrunk over the past
few years, with new technologies suddenly becoming “mainstream” more rapidly than ever before.
5.1
WiMAX (Worldwide Interoperability for
Microwave Access)
WiMAX is based on some of the technology from two failed
markets: LMDS (local multipoint distribution service) and
MMDS (multichannel multipoint distribution service.) LMDS
was intended to be a digital wireless transmission system in
the 28 GHz range in North America and 24-40 GHz elsewhere.
The purpose of LMDS was to replace wired CATV (community
access television), otherwise known as cable TV. For this purpose, LMDS acquired the oxymoronic title of wireless cable.
LMDS requires a clear line of sight between transmitter and
receiving antenna, which is from one to four miles apart,
depending on weather conditions. LMDS provides bandwidth
in the 51– 622 MHz range. This is considerably greater than
other wireless services, but is necessary for LMDS to accomplish its original task of replacing wired analog cable TV sys77
78
Unit 5: On the Bleeding Edge
tems. A few operating LMDS systems remain to satisfy some
rural customers, but satellite TV eventually has satisfied the
demands for the original wireless cable market.
MMDS was a second attempt to satisfy the need for wireless
cable, but at a more reasonable frequency band, in the 2.2-2.4
GHz range. It also requires a clear line of sight between transmitter and receiver, which can be thirty or more miles apart. It
was designed initially as a one-way service for bringing cable
TV to subscribers in remote areas or in locations in which it is
difficult to install cable. In late 1998, the FCC opened up the
technology for two-way transmission, enabling MMDS to provide data and Internet services to subscribers. MMDS too has
been displaced by satellite TV, but its two-way Internet access
has not been displaced.
WiMAX is intended for the general data communications market, which demands high bandwidth and highly reliable bidirectional interconnections. Sometimes this market is called a
backhaul in telephony terms; in data communications, it is usually referred to as a backbone network. The meaning is the same:
a network that serves to unite local networks into a larger single operational network. For this backbone task, WiMAX needs
to have high bandwidth, be resilient (the telecom word for
fault-tolerant), and must cover a large service area. However,
WiMAX is purely digital and was never intended to carry analog signals.
The task of defining a wide-area wireless network was given to
the IEEE 802.16 committee that created both the initial 802.16
standards document and its more recent extension, 802.16a.
The original task for 802.16 was to define high-speed wireless
services in the 2-66 MHz band. They defined the task so as to
cover two frequency ranges: 2-11 MHz and 10-66 MHz. IEEE
802.16a specifically addresses the 2-11 MHz band. Some of the
underlying technology for the standard was taken from
MMDS, since it has been proven to work. However, more than
simple broadband for television was intended for 802.16. It was
intended to supply all of the fixed (nonmobile) needs of a backbone network for voice, digital video, and data communications. IEEE 802.16 is a very complex standard with options for
Unit 5: On the Bleeding Edge
79
everything. The task of the WiMAX Forum is to develop profiles and test specifications for the many possible implementations of the standard.
The base data rate intended for WiMAX is specified as 268
Mbps. The standard specifies both FDM (frequency division
multiplexing) and TDD (time division multiplexing) for different data services. The initial standard, IEEE 802.16a is for fixed
(non-mobile) service, but the standard for mobile services,
IEEE 802.16e has now been completed. The data rate for mobile
services is somewhat lower than for fixed services.
WiMAX uses mesh networking to avoid the need for direct line
of sight between two points on the network. Each station automatically relays all messages not intended for itself to the rest
of the network. In the 2-11 GHz band, omnidirectional antennas are used. This enables the network to become quite large,
but with some multipath distortion as the signal is reflected
from buildings and other structures in the signal path. In the
10-66 GHz band, directional antennas are used to avoid multipath effects.
Cellular telephone carriers are very interested in using WiMAX
to connect their towers to central offices. Currently, these connections require expensive landlines, which are often rented
from the local exchange carriers (LECs), the existing wired telephone services. Those LECs that are also providing cellular services have a competitive advantage over other cellular
providers. A wireless backhaul would make cellular service
providers independent of the LECs, a financially and strategically important factor.
Internet service providers (ISPs) to business also have a strong
interest in WiMAX as a way of bypassing the need for expensive T1 or T3 service lines leased from the LECs. With WiMAX,
the ISPs can directly reach their customers and have better control over the performance of their broadband Internet services.
You can read more about the standard at: http://grouper.ieee.org/groups/802/16/.
80
5.2
Unit 5: On the Bleeding Edge
UWB (UltraWideBand)
In 2002, the Federal Communications Commission approved
the use of UWB for limited services including data communications. UWB is a new way to use radio transmission that consists of short pulses of low-energy radiation. The shape of the
pulse gives it the property of generating radio energy over a
wide frequency range, but at very low energy at any one frequency. The data is detected only by the presence (=1) or
absence (=0) of a pulse at the repetition time slot. This allows
UWB to overlap other radio bands such as Wi-Fi and the other
services in the 2.4 GHz ISM band without interfering. Generally, other radio modulation schemes such as DSSS will see
UWB as just impulse noise, which they easily filter out.
Since UWB uses pulses, it is capable of being detected over a
much longer range than other signal forms. Pulse signals also
tend to penetrate solid objects better than continuous wave signals. For example, one approved use of UWB is for groundpenetrating radar. This characteristic makes UWB an exciting
new technology with great potential for industrial automation
applications. Low-energy radiation requires less transmit
power and results in longer battery life for battery-powered
devices.
5.2.1
WiMedia
IEEE 802.15.3 was given the responsibility for high-data-rate
personal area networks (PANs). While many different air interfaces (radio) were considered, the committee decided that
UWB has the most potential. Initially, it appeared that the
pulse radio from the UWB pioneer XtremeSpectrum was the
only choice. Motorola considered this exciting enough to purchase that company. The original UWB technology was
referred-to as Direct Sequence (DS) UWB, and was favored by
manufacturers that banded together as the UWB Forum. However, a rival group called Multiband OFDM Alliance (MBOA)
challenged pulse radio technology with a more conventional
wideband radio that uses OFDM (orthogonal frequency division multiplexing) similar to that used in Wireless-A, G, and N.
Unit 5: On the Bleeding Edge
81
The WiMedia Alliance was formed to promote Multiband
ODFM technology that approximates the pulse modulation of
direct sequence UWB. However, the IEEE 802.15.3 committee
could not come to an agreement on how the two rival technologies could be written into the standard. In 2006, IEEE disbanded the 802.15.3 committee.
The WiMedia Alliance, now merged with the former MBOA,
found a new home for the base Multiband OFDM UWB standard in having it published as ECMA-368 that has also been
approved as ISO/IEC 26907. ECMA-369 (ISO/IEC 26908) specifies the low level interfaces.
WiMedia Alliance is the organization responsible for promoting the technology based on MBOA. The Alliance states its
objectives for WiMedia as follows:
•
High-throughput, wireless communications for multimedia
•
An easy-to-use, consumer-friendly solution
•
Based on international standards
WiMedia is intended primarily as a cable replacement technology for high-bandwidth applications such as streaming digital
video and for peripheral device attachment. While no industrial field applications obviously need WiMedia, it seems certain to find its way into configurations of control equipment to
replace cable. Cable replacement is usually well justified
because it reduces the cost of installation.
Meanwhile, future versions of Bluetooth will use WiMedia
technology and provide the needed upper layers of protocol
necessary for complete USB cable replacement for high speed
applications.
5.2.2
DS-UWB
IEEE 802.15.4a has recently been approved using DS-UWB as
its technology base. While the ZigBee Alliance has not made
any decisions on the use of this alternative PHY specification, it
82
Unit 5: On the Bleeding Edge
will be considered. One of the desirable properties of DS-UWB
is a higher data rate, up to 1 Mbps. Another property is location services; the ability to locate a transmitter within one
meter or less.
5.3
Wireless Sensor Networks
For the past ten years, there has been a vision for networkindependent smart sensors that are capable of using any convenient network technology. This vision has been based on work
begun by the National Institute of Science and Technology
(NIST). NIST has concentrated on the IEEE 1451 family of standards for connecting smart transducers to networks. The
microprocessor interface to sensors is called a smart transducer
interface module (STIM). Early work has produced IEEE
1451.2, a standard for Transducer Electronic Data Sheet (TEDS)
that specifies the digital interface for accessing that data sheet
so as to read sensors and set actuators. IEEE 1451 is not another
field network; it is an open standard that may be used with
multiple networks. IEEE 1451.1 describes the network-level,
object-oriented model of 1451 devices. The processing of the
sensor data is done in the network-capable application processor (NCAP), which is packaged with the smart sensor. The
devices described by IEEE 1451.1 and .2 are network independent. They have been demonstrated with direct Ethernet connections and with CAN networks as well.
IEEE 1451.4 moves the NCAP to the data acquisition device
(which is a plug-in board and computer, data logger, or standalone unit). The intention is to keep the size of the TEDS as small
as possible. 1451.4 defines a number of templates, which allows
a more compact representation of the data. The host must have
some understanding of the templates in order for it to decode
the TEDS information. With IEEE 1451.4, the network between
the NCAP and the sensor is defined as a very simple multidrop, serial communication protocol. This protocol requires
that a single master device (the system) supplies power and
initiates each transaction, with each node according to a
defined transaction timing sequence, on a single wire and
return. Dreams of wireless are just that — dreams.
Unit 5: On the Bleeding Edge
83
A new project, IEEE 1451.5 has now been formed to standardize on the movement to wireless sensor networks. Most of the
discussion has been about adaptation of Bluetooth technology
by creating a profile for wireless sensors. ZigBee has also made
its proposals to the working group. Since the 1451 parent committee is committed to avoiding the definition of any new network protocols, it is unlikely that a new network protocol will
emerge from this new effort.
The work on IEEE 1451 has been technically sound, but it has
not been adopted or supported by any industrial automation
suppliers. The TEDS concept originated in the EDS of ODVA
for DeviceNet and has been greatly expanded by FOUNDATION™ Fieldbus function blocks, Profibus EDDS, and HART
DDL. Now, these device descriptions have been unified by
their common inclusion into IEC 61804, Function blocks (FB) for
process control - Part 2: Specification of FB concept and Electronic
Device Description Language (EDDL), and ISA104.
The original vision of networks of “microwave-connected sensor chips” for collecting atmospheric and environmental data
has not been fully realized. It appears to be a wireless sensor
network that exhibits much of the character of ISA100: awakening on some schedule and broadcasting its value and status,
then returning to sleep to conserve stored energy (battery). To
cover a wide area, a mesh network topology, similar to that
defined by ISA100, may be necessary. It will probably not
require high data rates or long messages, and it will not require
node synchronization with any other node. It seems that the
dreamers of wireless sensor networks will experiment with
unique networks such as being supplied by Sensicast and Dust
Networks, but will eventually settle on ISA100 or something
similar.
5.4
Network Device Power
Wireless networks have existed primarily for portable or
mobile devices such as cell phones and other two-way radio
devices. The wireless LAN, WAN, and PAN have introduced
the new concept of wireless connections for its own sake — to
84
Unit 5: On the Bleeding Edge
eliminate the cost and inconvenience of wired connections.
With portable and mobile devices, battery power is a given,
since there is no other readily available power source. Soon,
this will change somewhat as fuel cells begin to offer alternatives to batteries for some devices. However, the subject of a
power source for wireless LAN, WAN, MAN, and PAN
devices has not often been discussed. Battery operation has
been assumed.
5.4.1
Optical
Significant power can be delivered over distances without wire
by using optical delivery methods. The most well-known optical power-delivery method uses photovoltaic cells, often called
solar cells, which are usually made from the following materials: single-crystal silicon, polycrystalline silicon, amorphous
silicon, and cadmium telluride. These materials are optimized
for generating electrical power from solar radiation, but they
are also used to power devices, such as calculators, through
incandescent and fluorescent light. Laser light may also be
used to deliver significant energy; however, high-power lasers
may be dangerous to humans and birds. Solar cells are often
used to power remote SCADA nodes.
The use of artificial light to power wireless sensors and other
automation equipment is not currently being offered, since
even today’s Bluetooth devices use too much power. The
essential ingredient for light-powered wireless field devices to
become practical is low-powered radio transmission such as is
being developed for ZigBee, WirelessHART, and ISA100.
Highly efficient, long-life light sources may be used in the
future to power wireless field devices as low-power wireless
protocols become accepted.
5.4.2
Pneumatic Power
You read it first here! Pneumatic power in the form of compressed air is required to operate many manufacturing processes and is still needed to operate the majority of process
control valves. It is therefore readily available in most manu-
Unit 5: On the Bleeding Edge
85
facturing facilities; 4-20 mA signaled pneumatic power is still
used in many instrumentation and control systems. The idea
here is to pipe the compressed air into wireless field devices
that are equipped with the ability to generate electricity from
the flow of compressed air. This can be done internally within
the device, or maybe in an external module. A very small
device would be needed to generate all of the energy required
for a low-powered wireless sensor and/or a control valve positioner. There are currently no products that are powered with
compressed air, nor are there after-market turbo-generators
suitable for powering remote wireless instrumentation.
5.4.3
Magnetic Induction
So far, the technology does not exist to send significant
amounts of power using a wireless method without endangering life. The only widely used method of wirelessly powering
remote devices is magnetic induction using low frequencies,
typically below 15 MHz. While component costs for magneticinduction power delivery are low, this technology is limited to
about three meters’ distance and very low power. While suitable for PAN usage, perhaps to power headphones, distances
appropriate for industrial automation LANs may not be suitable for magnetic induction.
AC electrical power can be induced from the AC power lines
that often run through process plants. The inductive coils can
be clipped over AC lines and can produce enough low power
DC to supply the necessary energy for low powered wireless
field instrumentation. Some small amount of field wiring is
necessary to pick up such power, but since most of this would
be low power circuits, it can be simplified and installed at low
cost. Currently, no devices are yet available to meet this need.
5.4.4
Microwave Power Transmission
NASA has long been interested in transmitting electrical
energy from solar collectors in stationary Earth orbit to ground
stations that would convert it back into electrical energy (see
http://www.seds.org/spaceviews/9608/nss-news.html).
86
Unit 5: On the Bleeding Edge
While the energy source would be solar, the transmission
would use a broadly spread beam of microwave radio. According to the SunSat Energy Council, a non-profit organization
affiliated with the United Nations, the beam would be so low
in density that it wouldn't even feel warm if you happened to
walk through it. While the success of this program is in doubt,
the technology for generating the broad microwave beam and
converting it into electrical energy has long been known. This
method may actually become practical when the power
required is only a few milliwatts, but currently there is no
known development of microwave power broadcasting for terrestrial applications.
5.4.5
Conversion of Waste Energy
Energy exists within manufacturing processes in the form of
vibration, thermal differences, flowing fluids, and often pressure differences. These sources of energy can be used to generate small amounts of electricity, perhaps enough to power lowenergy sensors with a wireless link. This technology is often
called energy harvesting or energy scavenging.
Millennial Net produces its I-Bean products as components
that can be used by product manufacturers. One of its recent
products uses an “energy harvesting” technology from Ferro
Solutions, Inc. in which ambient vibration is used to power the
communications interface, completely without batteries.
Unit 6:
Significant News for
Wireless Networking
6.1
Energy-harvesting Component Runs Wireless
Nets
Microstrain Inc. has devised an answer to “energy-harvesting”
with a component that can power wireless nodes directly from
ambient energy in the environment (see http://www.eetimes.com/story/OEG20031230S0004). The Williston, Vermont,
company recently received a $700,000 Small Business Innovative Research Grant from the Defense Department to develop
the technology.
Microstrain identified mechanical strain as the best source of
inherent energy as a result of rapid advances in the performance of piezoelectric materials. These materials change their
physical volume when placed in an electrical field or, conversely, generate an electrical field when subjected to mechanical strain. Not only is strain a commonly available force in
buildings and machines, but recent advances in piezoelectric
materials have made high-efficiency fibers commercially available. To extract sufficient electrical power from a strip of piezoelectric material bonded to a beam under variable stress,
Microstrain devised a power-management scheme based on
charge storage in a capacitor. The wireless circuit is held in the
off state until enough charge accumulates to drive it. It remains
to be seen if there is enough energy available even for circuits
such as ZigBee or ISA100, which are off 97 percent of the time.
6.2
Honeywell Introduces OneWireless Networks
Announced in June 2007, the OneWireless architecture is a selforganizing, secure, self-healing, mesh network designed to
87
88
Unit 6: Significant News for Wireless Networking
support multiple wireless protocols including ISA100.11a,
HART over wireless, and the Honeywell new line of XYR 6000
wireless field transmitters. OneWireless is designed to be the
only wireless network required for plant-wide applications. It
is designed to be the wireless network for all applications.
The OneWireless architecture is based on the use of either of
two different radios, both operating in the 2.4 GHz band: one is
the same as the radio specified for ISA100.11a using IEEE
802.15.4:2006, and the other is a narrowband radio designed for
frequency hopping similar in concept to the Bluetooth radio.
The former would be used in networks integrating ISA100.11a
devices from many suppliers, while the second would be used
where a system would not require integration with ISA100.11a.
Both networks may operate at the same time in the same area
without interference.
Additionally, OneWireless architecture uses IEEE 802.11 for an
optional wireless backbone mesh network with wireless field
routers. Typically, low powered wireless transmitters are not
required to perform the routing function, that being assigned
to the field routers. This architecture reduces or eliminates
slow mesh hopping and makes the wireless network more
responsive for closed loop control applications in the future.
Typically, the backbone routers are electrically powered
devices and do not need the battery conservation measures of
battery operated routers.
The previous wireless line, the XYR 5000 is still offered and will
work with the OneWireless universal gateway. The XYR 5000
instruments wirelessly transmit measurements to a base radio
networked to a control or data acquisition device like a recorder
or PC. Each base radio accepts the signals of up to 50 transmitters. The base radio is available with a choice of Modbus or
4-20mA analog signal output for flexible communications.
Honeywell’s XYR 5000 transmitters feature three to five year
battery life and a low-battery alarm. This line of wireless
instruments uses frequency hopping spread spectrum radio
modulation in the 915 MHz ISM band between the transmitters
and the base radio over distances up to about 600 meters.
Unit 6: Significant News for Wireless Networking
89
XYR 5000 instruments are available for gauge pressure, absolute pressure, vibration, and temperature. The rated accuracy is
±0.1% of full-scale reading at reference conditions.
See also: http://hpsweb.honeywell.com/Cultures/en-US/
Products/wireless/SecondGenerationWireless/default.htm.
6.3
Accutech Wireless Instrumentation
Adaptive Instruments offers its Accutech brand of pressure,
differential pressure, temperature, and acoustic (vibration)
transmitters using frequency hopping spread spectrum transmission in the 915 MHz ISM band to a base station. These are
battery powered devices with rated battery life up to five years.
The distance between the field transmitters and the base station
can be up to about 900 meters. The base station is wired to the
data acquisition network using Modbus over RS-485 or Modbus /TCP, and must be powered with 24 vDC or 120/240 vAC.
Accutech transmitters are rated at ± 0.1 % of full-scale reading
at reference conditions. The devices are rated for operation
from -40 to +85 degrees C, and are certified for intrinsic safety.
The base station is rated either NEMA 7 (explosion-proof) or
NEMA 4x (weather-tight).
See also: http://www.accutechinstruments.com
Unit 7:
Recommendations for
Wireless Networking
The industrial automation industry is ready for broad use of
wireless networks in process control and factory automation
systems. The Wi-Fi technologies are ready for adoption by
industry network consortiums, particularly for use where
Ethernet and the Internet Protocols are currently used. Unfortunately, standards for the device level wireless (ISA100.11a)
are not yet finally approved.
In most cases, Wi-Fi can be tried where one of the Ethernetbased networks such as FOUNDATION™ Fieldbus HSE, EtherNet/IP, Modbus/TCP, or PROFInet can be used. At this time,
considerable network planning will be necessary to make sure
that all nodes are within the actual radio range of today’s wireless access points. The investment should be made in only
Wireless-N devices, and of the commercial grade, not the less
expensive home grade. Wi-Fi networks have dead spots that
must be recognized during installation so that positions can be
adjusted to receive transmissions. External antennas can usually be used to move the reception zone without moving the
unit; only devices that have external antenna jacks should be
purchased. Often, the external antennas need to be directional,
which will extend the network to the limits required for the
installation.
The real payoff from wireless for industrial automation will
come when field devices having standardized wireless connections are available. Supplier proprietary specifications for their
wireless devices should be avoided unless there is a reasonable
migration path to ISA100.11a.
The wireless network technology tradeoffs are being evaluated
in the work to establish ISA100. To some extent, some of these
91
92
Unit 7: Recommendations for Wireless Networking
issues were evaluated to design WirelessHART, but against a
very limited application environment. In the opinion of this
author, unless there are unusual circumstances, the wireless
connection of HART instruments should be made only with
ISA100.11a conforming adapters, rather than using WirelessHART equivalents. WirelessHART is a single-purpose network and may cause some problems when installed in the
same area as ISA100.11a networks. Since ISA100.11a has been
designed as a universal network, it has the connectivity, flexibility, and security for integration of all field devices, including
HART.
At present, no organization is backing the adaptation of
WCDMA (3G) technology for industrial automation wireless
applications. Likewise, no organization is currently backing the
use of WiMAX for wireless LANs. Both of these technologies
are popular for use in digital voice networks, but have not yet
attracted attention of the consortiums responsible for industrial
networks, even though they exhibit most of the desired
features.
While the battles will usually occur in the physical (radio) and
data link layers (protocols), the best path for the user will be
defined when the supporting consortiums adapt their dominant wired industrial networks for wireless keeping their
established application layers: FOUNDATION™ Fieldbus,
DeviceNet, Modbus/TCP, EtherNet/IP, and PROFInet. The
application network that suits the industrial application is the
one to be selected. Although the consortiums sponsoring these
networks are not involved in selecting the “best” wireless technology for their markets, they are all actively engaged in the
ISA100 standardization effort. The Fieldbus Foundation has
committed to using ISA100.11a as the basis for the wireless
Fieldbus.
Unit 8:
Radio Frequency
Tagging
Although radio frequency (RF) tags have been around for over
16 years, they have only recently emerged into a unique market. The first application for RF tags was in the identification of
animals; tiny tags about the size of a grain of rice were inserted
under the skin of pigs. Readers, scattered about the pigpen,
record when the tagged animals came to the feed trough and to
identify individual animals for health records and at slaughter
time.
The latest, most prominent use of RF tags is in highway toll collection applications, where they are known as E-Z Pass, FastPass, etc. These active programmable tags allow vehicles to
pass through toll bridge and toll road collection stations without stopping, speeding up the collection process. Another
highly visible RF tag application, to identify persons purchasing gasoline at Mobile or Exxon stations uses the tiny passive
SpeedPass. Processing the identity of the person with a SpeedPass allows the gasoline purchase to be charged to the pass
owner’s credit card.
8.1
Types of Tags
Although RF tags are not identical, they have a common identification field, usually 64–128 bits in length and a unique
numerical value (see Table 5). They must have a source of electrical power in order to respond when queried by a reader. Different tag technologies are used to keep the tag’s cost at the
lowest possible levels for the intended application. Finally, tags
have readable memory, but which may or may not also be
writeable.
93
94
Unit 8: Radio Frequency Tagging
Table 5. Types of RF Tags
Radio
Frequencies
Used
Word
Reading
Power
Length,
Distance,
Source
Bits
Meters
ROM
138 KHz
13.85 MHz
64
Reader 0.04 – 3
EMF
ROM
13.85 MHz
64
Battery
EEPROM
138 KHz
13.85MHz
96, 128
Reader 0.04 – 3
EMF
RFID active
2, 3, 4 EEPROM
programmable
138 KHz
13.85 MHz
>128
Battery
3 – 10
>128
Battery
3 – 10
Battery
1 – 100
Type of Tag
EPC
Memory
Class Type
RFID passive
0
RFID active
4
RFID passive 1
programmable
Data tag
2, 3, 4 CMOS RAM 13.85 MHz
Flash RAM 985 MHz (UHF)
RF location
-
8.1.1
EEPROM or 303 MHz, 2.4/5.8 64
CMOS RAM GHz, UWB
3 – 10
RFID Passive Tags
The simplest tag is the Radio Frequency IDentification (RFID)
passive tag, which only supplies its own identity—the 64-bit
value—when read. During the manufacturing process, a different number that can never be changed is etched onto each of
these least expensive RFID tags. Power for the RF transmission
of the tag’s ID data field comes from the reader that emits a
low-energy electromagnetic field (EMF to power the tag. When
the tag is energized in this EMF radiation, it repeatedly transmits its own identity field value. A recent variation on this protocol allows the transmission to occur only once after a fixed
time delay, unless it detects a query message generated by the
reader. This variation allows many tags in the same field of the
reader to be read individually.
The EMF extends from the reader a distance dependent on the
form factor of the reader’s antenna. Some high gain antennas
focus the EMF to a narrow beam in order to increase the distance between tag and reader. Without a high gain antenna, the
reading distances tend to be rather short, typically 3-4 cm. With
Unit 8: Radio Frequency Tagging
95
a high gain antenna, reading distances of up to 3 meters are
common, but usually only a single tag can be read at a time. By
moving the high gain antenna, or moving the tagged objects
past the reader’s antenna, many tags can be read consecutively.
The passive RFID tag’s value is approximately the same as a
barcode’s; the tagged item is uniquely identified. The data for
the tagged item is usually located in a computer database, not
on the tag. Simplicity of the tag keeps the cost at a minimum.
However, the RFID tag needs only to be located within the
reader’s EMF field, not in a direct line-of-sight as required to
read a barcode. This feature enhances the RFID tag’s functionality for item-tracking over that of barcodes. To read RFID tags
not located in a predictable location, wide variations in the
reader’s antenna construct are often used. The most common
antenna for reading an RFID tag with a reader in a fixed location is the loop antenna, in which the tagged item passes inside
the loop. Imagine the loop as a portal or entrance in which the
antenna wire encircles the passageway, creating an EMF field
inside the loop. When the tagged item penetrates the EMF
field, the tag is read. Loop antennas are ideal for doorways and
conveyor belts to keep track of moving items. When a tag is
read, the reader creates a transaction record that is sent via a
network to a host computer where it is entered into the transaction database. With items in motion, as for a typical materialshandling operation, RFID tags provide the last known location
for each item.
The passive RFID tag, in the form of a credit card, is now being
used in automatic fare collection for public transportation systems. The card is similar to a magnetic stripe identity card but
does not need to have physical contact with the reader – only
be near the read station. The card identifies the pass holder and
allows passage at the fare collection point. More complex
designs exist for variable fare systems; the card is read twice,
once on entrance to the system and once on exit, billing the
user’s account for the calculated fare. Unlike some magnetic
stripe fare cards, there is no value retained on the RF card. The
value is contained only in a computer database, resulting in a
more reliable system.
96
8.1.2
Unit 8: Radio Frequency Tagging
RFID Active Tags
Active tags are powered by an on-board battery enabling
higher power transmissions to cover longer distances. Reading
an active tag involves the reader continuously polling to determine if any tags are within its reading range. Often, the reader
generates an EMF to signal that it is ready to read the active
tag. When the tag receives the poll read-request, it returns its
ID value. This pattern avoids wasting the active tag’s battery
life; the power required to receive is far less than that required
to transmit a signal. Due to the much larger reading range for
an active tag than for a passive tag, more than one tag is likely
to be within reading range at a time. A reading protocol usually exists to ensure only one tag is read at a time.
The most common uses of active RFID tags are in automatic
highway toll collection and in tracking railroad boxcars. Systems such as EZ-Pass and FastPass use active RFID tags that
can be read at distances of up to 10 meters when located
behind the windshield of an automobile. Readers are positioned above the lanes that are marked for use of the pass. As
the automobile equipped with the tag moves under the reader,
the tag is read and the ID number is identified with the tag
holder. For single toll positions such as for a toll bridge or tunnel, the tag holder’s account is debited immediately. For highway tolls, the entrance reading is saved for processing with the
subsequent exit reading, at which time the tag holder’s account
is debited for the calculated toll. The tag retains no data. The
use of active RFID tags has resulted in a 300 percent improvement in these automated tollbooths’ capacity compared with
manual toll collection.
In North America, all railroad freight cars are tagged with an
active RFID transponder as part of the Rail and Intermodal
Asset Tracking System. As the rail cars pass in front of readers
located at strategic rail switching yards scattered all over North
America, their identity is read and reported to a common tracking system. This allows the system to track the location of all
freight cars in North America to the specific switching yard.
This information is used in billing freight car usage and in
scheduling and routing freight cars.
Unit 8: Radio Frequency Tagging
8.1.3
97
RFID Programmable Tags
“Programmable tag” is an old name but means that the user
can write the ID number on the tag. Most often, the technology
used is the Electronically Erasable Programmable Read-Only
Memory (EEPROM) that can be written without removing the
chip from the circuit and erased without the use of ultraviolet
light. Depending on the chip used, these tags can be re-written
between 10,000 and 100,000 times. Write times are slow compared to computer memory circuits, so programmable tags are
not intended to carry any dynamic data. The identity field is
usually the same length as for factory written tags, 64-128-bits,
but sometimes this field can be written to represent 8 to 16
ASCII characters rather than a 64- or 128-bit binary number.
This may help identify items when using a handheld reader
since the ID can be interpreted more easily as a character field.
Both passive and active programmable tags are available,
although tag use will be replaced by more connected databases
using RFID tags or by more flexible RF Data tags.
8.1.4
RF Data Tags
RF Data tags can be readable and writeable. Typically there is
an ID field of the same 64-128-bit length as for both passive and
programmable RFID tags, but extensive read/write memory is
located on the tag as well. In all cases, RF Data tags are active
with a long-life battery on-board. They generally have a read
range equivalent to other active tags, up to 10 meters, depending upon the reader’s antenna gain. Writing distance is greatly
reduced, to about 3 meters. The read/write memory is usually
Flash RAM with capacity of up to 256 Mb. Flash memory is
organized into blocks similar to disk and is supported to maintain a file system like that used for disk drives. Except for the
wireless connection to an RF reader, RF Data tags are comparable to USB memory devices. Flash memory does not require
battery power to retain data. Read and write speeds for RF
Data tags with Flash RAM are similar to that of Universal
Serial Bus (USB) memory, but slower than disk drives.
98
Unit 8: Radio Frequency Tagging
A high performance RF Data tag may be produced using battery-powered CMOS RAM. Read/write times are similar to
that of computer main memory. However, the radio speeds
necessary to use this high-speed memory cannot currently be
achieved at low cost, at low power, or for long distances.
Although it does not require much power to retain memory in
CMOS RAM, any loss of battery power erases memory content.
Therefore, no commercial RF Data tags are available using
CMOS RAM.
8.1.5
Location Tags
Sometimes called beacon tags, location tags, proprietary VHF
radio devices, are attached to pallets or containers stored in a
large flat warehouse space. The tag generates an intermittent
signal with the tag’s ID value that is broadcast over the warehouse space. Usually, the tag is equipped with a motion sensor
so that the interval between broadcasts becomes shorter when
the tagged item is in motion. Readers are logistically located at
the corners of the warehouse space to receive the RF signals.
When a reader receives the beacon from the tag, it also receives
the strength of the signal, indicating the approximate distance
between the tag and the reader. In order to map the twodimensional location of the item in the warehouse space, the
beacon signal identification and signal strength must be simultaneously read by a second reader. Both readers then send this
data to warehouse management software that maps the tagged
item location in two-dimensional space using triangulation.
Current commercial location tagging technology locates the
tagged item to an accuracy of about 3 meters using two readers. Using more readers may increase the accuracy of location
but also extends the warehouse area.
Tags based on ultra wideband (UWB) communications are also
used in location service. In this system, all tags and readers
share a synchronized high-precision clock. The tag sends an
intermittent location signal with its ID and the time-stamp
from the tag’s clock. The receiver adds its own time-stamp to
the message providing the system software with a differential
time delay between the tag and the reader. This time delay is
Unit 8: Radio Frequency Tagging
99
proportional to the distance between tag and reader. Readings
from multiple readers then allow the tagged item to be located
to an accuracy of less than one meter.
A third type of location system uses the time delay between the
time that a reader sends its signal, and the time it receives an
echo of that signal as an indication of the distance between the
tag and the reader’s antenna to an accuracy of about 3 meters.
This system uses multiple antennas for each reader to reduce
the cost of the system.
It is theoretically possible to construct an RF tag with a Global
Positioning Satellite (GPS) receiver to transmit its location with
high accuracy. Here, the tag itself would compute location by
reception of signals from at least 3 GPS satellites. However,
such a tag would require an unobstructed view of the sky (outdoor use only), and the GPS circuitry would probably use too
much energy to be powered from a battery. No commercial
GPS tags are presently available.
8.2
Tag Encoding
As previously mentioned, early RFID tags were only encoded
with a 64-bit “license tag” number, and many tags are still so
encoded. However, when it became clear that some tags would
be used similar to barcodes—to identify manufactured items—
a standards committee was formed to establish a uniform formatting for tag ID data. The organizations responsible for the
successful establishment of Universal Product Code (UPC)
were the Uniform Code Council (UCC) in the United States
and European Article Numbering (EAN) International. The
organization responsible for the international establishment of
Electronic Product Codes (EPC) for RFID tags is EPCglobal, a
joint venture of the UCC and EAN International. RFID tag
numbering has, therefore, become an extension of EAN, the
international version of the UPC.
The complete organization for use of EPCglobal is called the
EPCglobal Network that consists of the following elements:
100
Unit 8: Radio Frequency Tagging
•
EPC – the electronic product code, a number designed
to uniquely identify a particular item in the supply
chain
•
ONS – the object name service, which tells the computer systems where to locate information on the network about the object carrying an EPC
•
PML – the physical markup language, which is used as
a common language in the EPCglobal Network to
define data on physical objects
•
Savant – a software system that manages and moves
information.
EPC is the successor to barcodes for product identification. Barcodes have limitations, such as
•
They require line-of-sight for scanning,
•
They have limited encoding capacity, and
•
They cannot receive and store information.
However, more than one million firms in more than 140 countries currently use barcodes across more than 23 industries.
Barcode labels are inexpensive to print, are often included in
product packaging at no incremental cost, and can be read by
all modern point-of-sale machines. Since barcodes will always
be less expensive to deploy than EPCs, both will coexist for
many years to come.
The EPC is a simple, compact “license plate” that uniquely
identifies objects (items, cases, pallets, locations, etc.) in the
supply chain. The EPC builds around a basic hierarchical idea
that can be used to express a wide variety of different, existing
numbering systems. EPC numbers can accommodate all EAN/
UCC keys, including Global Trade Identification Number
(GTIN), Serial Shipping Container Code (SSCC), Global Location Number (GLN), Global Returnable Asset Identifier
(GRAI), and Global Individual Asset Identifier (GIAI).
Unit 8: Radio Frequency Tagging
101
Like many current numbering schemes used in commerce, the
EPC is divided into numbers that identify the manufacturer
and product type. In addition, the EPC uses an extra set of digits, a serial number, to identify unique items. The EPC is the
key to the information on its associated product that exists in
the EPCglobal Network. An EPC number contains the following items:
•
Header - identifies the length, type, structure, version
and generation of EPC
•
EPC Manager - identifies the company or company
entity
•
Object Class - similar to a stock keeping unit or SKU
•
Serial Number - specific instance of the Object Class
being tagged
Additional fields may also be used as part of the EPC to properly encode and decode information from different numbering
systems into their native (human-readable) forms.
EPC Manager numbers, issued by EPCglobal, are required for
companies that engage with trading partners outside of the
four walls of their internal operations. EPCglobal Networkcompliant software and hardware utilize EPCglobal standard
data protocols, thus requiring the use of an EPC Manager
number.
8.2.1
EPC Global Gen2 tags
In late 2004, the EPC Global Gen2 tag specification was
approved. At the same time, the EPC Global organization
agreed to provide no further conformance testing for the older
Gen0 and Gen1 tags, but to devote its efforts to interoperability
testing for Gen2 tags. The specific nature of the approved Gen2
tags includes an air interface in the 985-MHz UHF band and
requires an ID field of at least 96 bits. Both active and passive
tags may be used. The protocol provides for multi-tagging,
allowing separate reading of multiple tags in the same reader
102
Unit 8: Radio Frequency Tagging
field. The specifications are not yet published for the general
public but are being drafted for ISO submittal for international
standardization.
8.3
Alternative RFID Standards
EPCglobal is considered to be a North American and European
standard, with little participation from Asia. Both Japan and
China have their own standards in this area.
•
UID – Ubiquitous ID, the Japanese RFID standard
•
NPC – National Product Codes, China’s RFID standard
Since Japan and China strongly depend on their exports to
North America and Europe, it is unlikely that their regional
standards will do more than influence future revisions of EPCglobal, except in their local home markets.
8.4
RF Database Tag
RF Database tags are not yet in wide commercial use. The tag
type is the RF Data tag described in Section 7.1.4, but the application is to hold a database designed to travel between
domains. Eventually, this type of tag will be widely used to
solve a very difficult problem in the integration of data
between companies on both sides of the supply chain.
Simple RFID tags contain a unique number used to identify the
tag. To identity the tagged item, the tag number must be associated with the item’s identity such as its Stock Keeping Unit
(SKU). That association is found in a database. When the goods
are shipped, the portion of the database describing the shipment, including the RFID tag number, is sent from the manufacturer or distribution center (DC) to the customer. For the
customer to use the database, the format must be known. In the
past, the format was defined, in many cases, by a standard
known as Electronic Data Interchange (EDI), a complex encoding based upon use of Abstract Syntax Notation (ASN.1), an
international standard for defining data structures. Data formats for interchanging data were defined for many industries
Unit 8: Radio Frequency Tagging
103
using the standard identified by the ANSI X.12 committee. To
use EDI, a private communications link between trading partners was required, limiting use to only large companies. There
was also the need for consultants capable of using ASN.1 to
specify the database structures and to adapt these structures
for model and usage changes.
ANSI X.12 has now been replaced by Electronic Data Interchange For Administration Commerce and Transport (EDIFACT), defined by the ISO 9735 standard, and administered by
the United Nations Economic Commission for Europe
(UNECE). Under direction of UNECE, new standards have
been prepared for building data exchange standards in many
industries and applications. The latest efforts for simplifying
EDIFACT use eXtensible Markup Language (XML), an Internet
standard. XML uses human-readable data identifiers to define
data items and structures.
The primary problem in data handling across the supply chain
is associated with timing. Often, the goods are delivered before
the old rules of EDI, or even the newer EDIFACT formats, can
deliver the data. Therefore, the goods remain on the receiving
dock until they are registered in the on-line inventory database. Initially, this problem was to be corrected using barcoded
shipping container labels, but these required special equipment to read the two-dimensional barcodes, operator training
to read all of the item barcodes on the shipping label, and the
discipline to actually read the label when the goods were
received. RF Database tags are designed to solve this problem
in real-time without the use of real-time data communications,
specialized training, or difficult-to-administer work processes.
An RF Database tag reader/writer records the shipping container information when the container is loaded to the common
carrier at the manufacturer or DC. Another RF Database tag
reader reads the RF Database tag as the shipping container is
received from the common carrier and then registers the data
as a transaction with the on-line database. For factory operations using just-in-time processes, another layer of work-inprocess inventory can often be removed, immediately paying
for the equipment installed.
104
Unit 8: Radio Frequency Tagging
RF Database tags can also be used to transport item data during the manufacturing process. While Information Technology
(IT) networks are widely deployed to the shop floor in large
assembly line manufacturing operations, they are not usually
designed to deliver real-time data to the workstations where
parts are machined or assembled. For example, when an automobile chassis is moved into a workstation, the IT network
may take an indeterminate time to look up the chassis ID verified by a barcode scan and inform the operator which of several options are to be mounted. This information, including full
text descriptions of the options, can be contained on an RF
Database tag for immediate action by the manufacturing cell
workstation. Furthermore, the ID of the options can be added
to the tag, along with any quality test results as manufacturing
or assembly takes place, keeping a running record of the
assembly with the chassis. The RF Database tag creates a traveling database that functions within the manufacturing environment without requiring a real-time network to be installed.
8.5
RF Tag Recommendations
Due to the requirements of significant users such as Wal-Mart
and the US Department of Defense (DoD), RFID tags using
EPCglobal standards will be widely used. These tags are basic
identity for products and will be expected for inventory control
as well as all transaction processing. The investment in RFID
for control of work-in-process, shipping, and receiving inventory will also be highly beneficial in industries not directly
affected by Wal-Mart or the DoD, and in which previous studies of barcode use did not prove to be effective. Contrasted to
the human actions required to read barcodes, the automatic
reading of RFID tags will yield positive cost saving results.
Barcoded shipping container labels are supplied in many
industries, especially automotive parts manufacturing and
automobile assembly. However, in some industries, shipping
container barcode labels are less than effective due to the manual scanning step required before data can be used. RF Database tags can overcome this problem with properly configured
readers. Inventory reduction resulting in a net positive cash
Unit 8: Radio Frequency Tagging
105
flow benefit should ensure that RF Database tags become popular. However, as the industry has not focused on this area of
application, RF Database tags are still relatively expensive.
To replace shipping container barcodes, RF Database tags must
follow industry standards for encoding the contained database.
Using the EDIFACT encoding methods based on XML is the
key to standards; however, this work is incomplete. Many
years of development led to the X.12 EDI standards for each
industry, and now much of these efforts are being converted to
EDIFACT standards. When these standards are complete, they
will become the benchmark for each shipping container
description.
About the Author
I am now CEO of CMC Associates
(Acton, Massachusetts), which is to
say that I am an independent consultant and can give myself any title I
want. I have been actively involved in
industrial automation work since 1958
when I started doing instrumentation
for a small chemical plant of Ethyl
Corp. in Baton Rouge, Louisiana. Not
too long after graduating from the
University of Florida in chemical engineering, I began working on my masters in science at Louisiana State
University in Baton Rouge. Paul Murrill and Cecil Smith, now fellow ISA authors, were in my graduate automatic control class, the first ever taught at LSU. In
1964, I received my M.S. in chemical engineering.
In 1964, I became one of the pioneers in computer control while
working at Union-Camp in Savannah, Georgia (now part of
International Paper). There I developed, installed, and operated an IBM 1800 computer for control of both a fast Kraft
paper machine and a Kamyr continuous digester, and consulted to the Franklin, Virginia mill for bleach plant control. I
performed all the software design and FORTRAN programming for this real-time system, which actually performed
closed-loop advanced feedback control. To think that the IBM
1800 had less computing capability and disk storage than the
very first IBM PC fifteen years later! I feel very fortunate to
have had the chance to be a control systems pioneer.
Foxboro Company was my next stop. I went to work immediately on its PDP-8-based control systems. I led the team that
converted its control systems to the PDP-11 as the FOX/2 and
2A. Later, I led the team that brought the FOX/1 to market in
my first project as a department manager. I then became the
107
108
About the Author
marketing guy for Foxboro’s computer control products and
planned the successor line, the FOX/1A. My final assignment
at Foxboro was in the R&D area, where I ran a project to introduce a new architecture into control systems. Along the way, I
earned my MBA.
With computer control as my specialty, I was recruited by Ken
Harple, the founder of ModComp in Ft. Lauderdale, Florida,
my hometown, as ModComp’s Director of Industry Marketing.
ModComp needed control systems software, so I worked with
my old friend Cecil Smith to create a control systems package
for its computers. After a financial meltdown at ModComp, I
found myself working for Cecil, selling his software on ModComp and other computers.
Following this, Ken Harple again recruited me for Autech Data
Systems, a company he had formed to build process control
systems after he had been forced out of ModComp. This was
great fun and gave me the chance to design my own DCS, the
DAC-6000, and a Faultproof system. It was the first DCS to feature ruggedized fault-tolerant controllers, an Ethernet-based
fiber-optic network, and a PC-based touchscreen operator console, all exhibited at ISA 1983. In this same period, I joined the
ISA SP50 standards committee to help develop Fieldbus. Failure to secure financing forced Autech to shut down before we
could become self-sustaining.
After solving problems for Computer Products, Inc., Analogic,
and other companies as an independent consultant, I moved
back to Massachusetts to work for Arthur D. Little, Inc., (ADL)
a world-class technology-based consulting company. ADL
taught me the dynamics of consulting. Most of my time was
spent in new product innovation and telecommunications, but
I also did some industrial automation. One of my projects was
to design the mechanism for detecting and suppressing commercials while recording video broadcasts so the VCR can fastforward past commercials without missing any story material.
This innovative project resulted in two U.S. patents: 5,455,630
in 1995 and 5,692,093 in 1996. ADL sold licenses for it to all
VCR manufacturers as Commercial Advance™. In this period, I
About the Author
109
also took over the management of the ISA and IEC fieldbus
standards committees.
When ADL began the downward spiral that eventually led to
its bankruptcy and dissolution, I joined Andy Chatha at ARC
Advisory Group. ARC gave me a marvelous platform to influence the automation industry. During this time, the international fieldbus standard was completed and published. ARC
gave me the opportunity to spread the word on the use of
Ethernet for industrial automation, initiating the trend I actually began in 1983, toward its widespread use today.
Now, at my own consulting company, I have the chance to
help many companies, but at a more leisurely pace. Writing
books was not my chosen profession, but it is an honorable one
and certainly fills my days. These days my time is filled with
work on the emerging ISA100.11a standard, where I co-chair
the User Working Group. Building consensus among the users
and resolving differences from WirelessHART is intellectually
stimulating and a valuable service.
I have been privileged to receive a number of awards, mostly
from ISA. In 2000 I received the ISA Standards Award for my
leadership in completing the ISA 50.02 and IEC 61158 fieldbus
standards. In 2001 I was elected to be an ISA Fellow, a lofty
honor indeed. In 2005 I was elected to the Process Automation
Hall of Fame. Also in 2005 I became an ISA Certified Automation Professional.
Richard H. Caro, CEO
CMC Associates
2 Beth Circle
Acton, MA 01720-3407 USA
[email protected]