Besides giving the Keynote address at EMC+SIPI 2017, Julius Knapp also played keyboards at the welcome reception.
National Harbor, Maryland — "Design IoT devices for immunity now," said Julius "Julie" Knapp, Chief of the FCC's Office of Engineering and Technology, at the 2017 IEEE Symposium on EMC and Signal/Power Integrity keynote address. "Design now for the unexpected."
By "unexpected," Knapp referred to the wide range of frequency bands that connected devices might use. Connected IoT devices use both licensed bands such as those assigned to wireless carriers as well as unlicensed bands, such as those used by Wi-Fi and Bluetooth. The box below (from Knapp's slide) highlights the spectrum available for wireless IoT devices.
- Existing commercial wireless bands allow flexible use
- Expansion of Spectrum:
- AWS-3 Auctioned 11/14/14
- AWS-4 — Mobile Satellite S-Band spectrum to terrestrial
- TV incentive auction — 600 MHz
- 3.5 GHz — Advanced sharing (Priority Access Licenses)
New licensed bands in millimeter-wave spectrum at 28 GHz, 37 GHz, and 39 GHz
- Existing unlicensed bands allow flexible use:
- 915 MHz: (902–928 MHz)
- 2.4 GHz (2,400–2,483 MHz)
- 5 GHz (Total of 555 MHz)
- 57–64 GHz (7 GHz)
- Overlay in many other bands
- Expansion of unlicensed:
- New band at 64–71 GHz
- White spaces in TV and 600-MHz bands
- 3.5 GHz — Advanced sharing (General Authorized Access)
- Relaxed existing 5-GHz rules
- Considering additional spectrum at 5 GHz
Why design for immunity? Because a band adjacent to the band in use today may be relatively unused and quiet, but that doesn't mean the adjacent band won't be used more extensively in the future. If a wireless device doesn't sufficiently reject signals from adjacent bands, that energy could interfere with a device's operation. As spectrum allocation changes, interference from adjacent signals could cause failures. "Assume adjacent bands will have more activity in the future," explained Knapp when EE Times spoke with him following his address.
Knapp, in his address, also touched on the FCC's role regarding IoT devices. While the FCC assigns the frequencies used by devices and systems in the U.S., the agency doesn't set the technical standards for the actual communications signals, protocols, or interoperability. That's in the hands of standards bodies such as IEEE.
When designing for immunity, Knapp suggested that engineers look to the many immunity standards in place for the wireless industry as well as EMC immunity standards. "Receiver bandwidths tend to be wider than transmitter bandwidths," he said to EE Times. "Keep receiver bandwidths in line with those of transmitters and design to immunity standards."
In his address, Knapp also mentioned potential problems with an IoT device's software-controlled radio. Today's devices can change frequency and power based in software controls. Software patches must not cause a device to go out of compliance with technical standards.
Many IoT devices are being developed by "makers" and startup companies. Many designs incorporate RF modules that add wireless connectivity. The use of such modules is pushing the RF compliance up the supply chain to the module makers, but ultimately, a finished connected device must still operate within its environment and the manufacturer is responsible. Then there's the issue of devices coming from Asia. When asked if businesses selling extremely low-cost devices might not make their devices compliant with standards, Knapp explained that radios must be certified and registered in a database. "We can take action if non-certified devices are being sold," he said.
—Martin Rowe covers test and measurement for EE Times and EDN. Contact him at firstname.lastname@example.org