The European Telecommunications Standards Institute (ETSI) recently conducted a test where it outfitted all of the dock doors in a Metro distribution center with passive UHF interrogators based on the second-generation Electronic Product Code air interface standard. It ran pallets of product through the dock doors simultaneously to see whether all the pallet tags could be read, or if the interrogators would interfere with each other (see ETSI Tests Show EPC Scaleable in Europe). The test was a success, but I was skeptical.
Some people I spoke to at the EPCglobal Conference in Los Angeles suggested the test wasn’t exactly rigged, but nonetheless was not a good test of real-world conditions. Others have told me for more than a year that the recently adopted ETSI regulations for RFID interrogators in Europe just don’t provide enough bandwidth for systems to work efficiently.
I hosted a panel during RFID Journal LIVE! Europe last week in Amsterdam on the topic of the performance of ETSI-compliant UHF interrogators. On the panel were Christian Plenge, RFID project lead for Metro Group International, which has tested ETSI-compliant interrogators from a variety of manufacturers, and Josef Preishuber-Pfluegl, co-chair of an ad hoc advisory committee set up by EPCglobal’s Hardware Action Group to study issues surrounding the performance of ETSI-compliant interrogators. He is also a member ETSI’s Task Group 34, which devised the new regulations, and chief technology officer of CISC Semiconductor.
Preishuber-Pfluegl, who helped to design the tests that ETSI ran at the Metro DC, said they were based on input from EPCglobal subscribers. “We ran the pallets through the dock doors simultaneously, which would not happen in the real world,” he said. “So this was a worst-case scenario.” The testers also created RF interference to see how well the interrogators would work in a noisy RF environment, and still all the tags were read. Plenge confirmed that the tests were legit and the performance of ETSI-compliant UHF EPC interrogators was good—even better than he had expected.
Some of the expected problems with UHF ETSI-compliant interrogators centr on something called “listen before talk.” Under the old ETSI rules, UHF interrogators could only operate 10 percent of the time in order to allow other devices to use channels in that area of the spectrum. The new rules removed the 10-percent restriction on “duty cycle,” but require interrogators to listen on a channel before using it. After emitting energy for 4 seconds, they must stop and listen again before using the channel again or switching to a new channel.
The concern is that the threshold for picking up an RF signal on a channel is so low (96 decibels) that interrogators might not be able to function when there are a lot of them in a confined space. There might be so much ambient RF energy in the facility, even after interrogators stop emitting energy, that they would not function properly. The tests confirmed that this is not a problem, and both Plenge and Preishuber-Pfluegl confirmed that this is, in fact, the case.
I spoke to a woman who works for another major retailer in Europe that is adopting EPC technology. I don’t want to use her name or her company’s name because we spoke informally, but she said there could be issues with ETSI’s listen-before-talk requirement. “At the sensitivity levels ETSI has established, you can detect another reader on the channel from a kilometer and a half [5,000 feet] away,” she said.
This woman went on to say that when you have a steel-framed dock door and a forklift truck with metal skin, you create a sealed RF environment that makes it easy to operate interrogators in fairly close proximity. “But some of our older buildings have brick walls, and we don’t use forklift trucks with metal skin,” she said, “so in some cases, we have a completely open RF environment.”
Still, she indicated that the new ETSI rules seem to be sufficient to enabling interrogators to operate without interfering with each other. The LBT requirement works well when interrogators are facing one another. Her one question was whether it would work as well when RF is leaking from interrogators lined up side to side.
That problem could go away if ETSI adopts a new proposal, which would have four channels in Europe dedicated to RFID. These channels would not require interrogators to use the LBT protocol. Other short-range RF devices currently using that area of the spectrum would move to the current band where RFID interrogators would still be required to use LBT. These new regulations could be adopted next year.