A Fatal Exception Has Occured In Your White Spaces Sensing Device

It would be funny were it not so easy for NAB to exploit.

The Microsoft prototype shut itself down last week and would not restart. Users familiar with MS products that are scheduled for release, never mind pre-beta versions, will find this so unremarkable as to wonder at the sensation. It goes up there with “Apple denies latest i-rumor.”

Unsurprisingly however, the folks opposed to the use of white spaces (primarily the broadcasters and the wireless microphone folks, with a dash of the cable folks thrown in for good measure), will spin this as the entire technology for sensing if a channel is occupied as “failing.” This ignores the other prototypes of course (Phillips and Google), and ignores the fact that the failure had nothing to do with the sensing (the thing being tested). Finally, of course, it ignores the fact that this is a proof of concept prototype.

The fact is, that the FCC testing shows that “sensing” as a technology works at levels that easily detect operating television channels and even wireless microphones. In fact, it is too bloody sensitive. In a foolish effort to appease the unappeasable, the companies submitting prototypes keep pushing the level of sensitivity to the point where the biggest problem in recent rounds appears to be “false positives.” i.e., it is treating adjacent channels as “occupied.”

As a proof of concept, that should be a success. The testing demonstrates that you can detect signals well below the threshold needed to protect existing licensees. Logically, the next step would be to determine the appropriate level of sensitivity to accurately protect services, set rules, and move on to actual device certification based on a description of a real device.

But that is not how it works in NAB-spin land. Instead, NAB keeps moving the bar and inventing all sorts of new tests for the devices to “fail.” For example, the initial Public Notice called for prototypes for “laboratory testing.” MS and Phillips submitted prototypes that performed 100% in the lab. But then, the MS people did something very foolish, but very typical — they decided their laboratory device was good enough for field testing. No surprise, it did not work as well in the field as in the lab. As this was a laboratory prototype, the failure to perform flawlessly in the field should have been a shrug — it would have been astounding beyond belief if a prototype designed for the lab had worked perfectly the first time in the field. But the fact that the prototype did not work in the field was widely declared a “failure” by NAB, which unsurprisingly gave itself lots of free advertising time to spin the results this way.

So the FCC went to round two, and again the NAB and white spaces opponents have managed to move the bar so they can again declare a “failure.” Back in 2004, when the FCC first proposed opening the white spaces to unlicensed use, it concluded that operation of white spaces devices would not interfere with licensed wireless microphone users. The FCC has never reversed that determination. Unsurprisingly, businesses developing prototypes according to the FCC’s proposed rules have not taken particular care to address wireless microphones. Because the FCC explicitly said “don’t worry about them.”

But suddenly, if the devices can’t accurately sense and detect wireless microphones, they will be “failures.” It doesn’t matter that the devices have proven they can protect wireless microphones. It doesn’t matter that Google has proposed additional ways of protecting wireless microphones besides sensing. As long as NAB can frame what defines “failure” (rest assured, there will never be any successes of NAB gets to call the tune), and can keep changing that definition at will, the political environment will ensure that the actual engineering is irrelevant.

Which is why the companies need to stop trying to placate the NAB by agreeing to an endless series of tests with ever-shifting criteria. And OET needs to write up a report that does what the initial notices promised to do, use the data collected from prototypes to determine if the concept works and, if so, to set appropriate technical standards. The prototypes have proven they can detect signals with a sensitivity better than an actual digital television set or wireless microphone receiver, so the “proof of concept” aspect stands proven. Rather than buy NAB spin, the next step should be to determine what level of sensitivity to set as the standard.

Hopefully, the Office of Engineering and Technology, which is conducting the tests, will not suffer the fate of the Microsoft prototype and shut down under pressure.

Stay tuned . . . .

Be Sociable, Share!
This entry was posted in Spectrum, Tales of the Sausage Factory and tagged , , , , , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • Connect With Us

    Follow Wetmachine on Twitter!

Username
Password

If you do not have an account: Register