Wireless Standards Then and Now

May 14, 2012
Industrial Networking Technologies Have Advanced, but the Industry is Still Dealing with Many of the Same Past Issues
About the Author
Aaron Hand is Managing Editor of Control Design and Industrial Networking. He joined Putman Media recently after almost 20 years covering high-tech industries, including semiconductor, photovoltaics and related manufacturing technologies.As Executive Editor Jim Montague reflected in this space in our Q1 issue, Industrial Networking celebrates its 10th anniversary this year. He took us on an interesting ride that showed that, while industrial networking technologies have advanced over the past decade, industry is still dealing with many of the same issues today as it was when we launched our first issue.A deeper dive into wireless standards tells a similar story. The technology has progressed, certainly. The concerns about reliability and security are not at the same levels they were 10 years ago, but users are still reluctant to put too much trust in wireless technologies.

When Dan Hebert, then technical editor, wrote a cover story on wireless industrial networks for our Fall 2002 issue, the message from the user community was clear: "Provide workable, supportable, non-proprietary solutions based primarily around wireless Ethernet and its 802.11b extension, and maybe even Bluetooth." But wireless standards are still duking it out for market dominance.

A decade ago, as Hebert noted, reliability concerns were at the forefront in a society where cellphones failed 5–10% of the time and cordless phones still didn’t hold a candle to wired reliability. Today, however, many people have dropped their landlines altogether, relying almost entirely on their cellphones for their primary communication. But making wireless a more accepted standard for industrial control will require continued vigilance and development within the reliability realm.

Over the years, wireless networking has gone through its security ups and downs—what security guru Eric Byres in a series of columns beginning in Spring 2003 referred to as "the roller coaster ride of IEEE 802.11b Wireless Ethernet." As Byres notes, when the specification was drafted, it was intended to provide a level of security similar to wired standards, which were relatively limited at the time.

Security continues to be a primary concern for industrial environments, but security has advanced. Today, both WirelessHART and ISA-100.11a standards provide network IDs in a secure environment before connecting the device to the wireless network.

For example, HART gateways and devices must have a physical connection to the network (through a modem) in order to operate on the network.

End-to-end sessions using AES-128 bit encryption ensure that messages can be deciphered only by the final destination.

After being approved earlier in 2011 as an ISA standard, ISA-100.11a-2011 was approved in September by the International Electrotechnical Commission (IEC) as a publicly available specification (PAS), setting it on the road to more global acceptance.

The argument supporting the ISA wireless standard is that it was developed through an open-consensus process and accredited by the American National Standards Institute (ANSI). "From my perspective of 42 years as an end user in the refining and petrochemical industry, I am especially proud that ISA-100.11a is the first industrial wireless standard developed in an open, accredited standards process with direct input and participation of experts from end-user companies," said Herman Storey, ISA100 co-chair, in reference to the September IEC approval.

Even with increasing acceptance, wireless networking is unlikely to be accepted anytime soon for critical control applications. ISA-100.11a-2011 was developed for non-critical monitoring, alerting, supervisory control, open- and closed-loop control applications. The standard defines specifications for wireless connectivity for applications with low data rates and very limited power consumption requirements. Applications, such as monitoring and process control, should be able to tolerate latencies on the order of 100 ms.

The HART Communication Foundation, however, insists that—despite all the naysayers in publications, forums and blogs—laboratory tests with WirelessHART (which is also IEC-approved) contradict concerns about determinism, reliability and security. A technical note from the organization contends that control performance of a typical WirelessHART network is comparable to that of traditional wired fieldbuses. "The WirelessHART protocol allows for secure, highly reliable, low-latency control with almost no impact on the bandwidth and absolutely no impact on process performance."

Nonetheless, even HART's own director of technology programs, Ed Ladd, recommends sticking with non-critical control applications.