After analog TV stations go off the air, there will be even more white spaces
For several years now, the FCC has been pursuing the possibility of allowing unlicensed data transceivers on TV frequencies, in the channels that aren’t being used, also known as “white spaces.” While the primary focus has been the threat of interference to broadcast TV stations and the wireless microphones they use, the cable industry needs to be aware of the interference threat from these devices – because for cable systems, there are no white spaces. And TV receivers are notoriously susceptible to “direct pickup” interference; that is, interfering signals entering the TV through its chassis.
Starting in 2004, in response to advocates like Google and Microsoft, the FCC has been on a path heading toward opening up a potentially huge amount of spectrum for unlicensed devices, at least in some parts of the country. In any city, the TV broadcast spectrum carries some TV channels, but most of the 6 MHz channels are vacant. There are great distance separations between two broadcasters who use the same channel, in order to avoid interference from one city to another. After analog TV stations go off the air next year, there will be even more white spaces.
The theory is that these white spaces can be used, and devices can be designed so that they select frequencies that don’t interfere with existing spectrum users. It’s an interesting theory, but not yet shown to be valid.
So work has been underway to develop technical rules that will prevent these unlicensed devices from causing interference. One approach is to require a device to contain a GPS receiver so that it knows its own location, and a database of TV station transmitter locations. That might work to protect TV receivers. But there are many broadcasters who use wireless microphones and intercoms that operate on the TV frequencies – for example, during production of a baseball game. Those itinerant uses don’t appear in any database.
Another approach is “listen before transmit.” The unlicensed devices would scan the allowed frequency range and find those channels that are not being used. Test results of prototypes have not yet confirmed that this works. Either devices did not detect signals that were there, or they thought they detected signals that were not there. And since wireless mikes put out very low power signals, it is not clear that the unlicensed devices could detect wireless mikes.
And then there is the beacon approach, where wireless mikes would add an easily-detectable signal on a known frequency. But adding that to the multitude of deployed wireless mikes seems like a challenge.
The FCC has a continuing series of tests underway of prototype devices, from companies such as Motorola, Philips, Microsoft and Adaptrum. For example, the FCC Labs recently did bench tests of devices for sensing in the presence of DTV signals in adjacent channels. Eventually they will include field tests, including (maybe) interference to cable TV reception.
The FCC’s test plan says that, if the FCC picks a test site where cable service happens to be available, a test to evaluate direct pick-up interference will be performed to the extent that it is feasible to do so. The device transmitter will be tuned to a receivable cable channel and activated at various locations while observing the television picture quality for signs of degradation. Not exactly the most detailed test plan.
Actually, the FCC has already done one test of direct pickup interference by digital TV receivers, and it failed miserably. In a report released last summer, which received almost no notice, the FCC said that it tested three DTV receivers and they experienced direct pickup interference at transmitter levels far below what is being contemplated for these white spaces devices.
The FCC proposed permitting fixed white space devices to operate at EIRP levels up to 36 dBm (1 watt power with up to 6 dBi antenna gain), and an EIRP limit of 26 dBm (100 mW power with up to 6 dBi antenna gain) for portable devices. These levels are comparable to the limits for WiFi routers – they are allowed to use up to 36 dBm, but in fact, most operate at or below the 26 dBm level.
So here’s the punchline. The FCC found that with a 2 meter separation including a wall between the unlicensed device and cable-ready DTV receiver, a transmission as low as 8.5 dBm produced interference. With a 10 meter separation and a wall, as little as 15.4 dBm produced interference.
The FCC tested 108 configurations, with transmissions entering the TV chassis through the front and through the back. They tested different types of walls. So there was a fair amount of variability in test results. Most of the tests were done with rear entry, as might be caused by the neighboring apartment, and the median interference threshold EIRP was found to be 16.9 dBm at 2 meters and 24.2 dBm at 10 meters, still way below the proposed limit.
Direct pickup interference has always been a problem for TV receivers, and the solution has always been to supply set-top boxes, which are far more robust. In spite of the shift to digital, some things never change.