Cognitive Radio

There is a cool trend happening where our wireless interfaces are getting smarter and smarter. This has been coined Cognitive Radio (this is related to Software Defined Radio and Dynamic Spectrum Access Networks ).

A recent highlight from MIT Technology Review's emerging technologies takes us to some research being done in UC Santa Barbara by Heather Zeng on the topic.

I was first introduced to this about a year ago on a campus visit to the University of Colorado at Boulder. One of the profs there, Douglas Sicker had been working on the SDR side of things for a bit of time and appears to still be working in that direction.

It hit me again last fall in a talk given by Benny Bing on the topic of "Broadband Wireless Access - The Next Wireless Revolution". Here is one of his slides as an example.

My understanding is fairly limited, but the idea is to move the radio and management of the radio further up the stack allow software to modify and shape the radio output (modulation, demodulation, signal processing, etc.). This allows really nifty things like frequency and power shifting to the software layer based on feedback and monitoring of the spectrum. So SDR refers to the radio side of things, better/smarter hardware, while Cognitive Radio is the software that takes advantage of this shape shifting hardware and is taking some shape in the 802.22 standard. Wikipedia offers some good info on Cognitive Radio including the claim that the term was first officially presented by Joseph Mitolla III and Gerald Q. Maguire, Jr. in an IEEE article in 1999.

Of course this represents some interesting challenges for the policy, certification and enforcement sides of things. It also has some interesting challenges on the technical side, which is what the MIT article represents -- the use of game theory to determine what spectrum to use...

This certainly has impact on urban zones where spectrum is heavily used, especially with the release (at some point) of the existing TV bandwidth as we move to HD. I'm wondering if this helps in rural and developing areas. I can see advantages as far as being able to make equipment compliant to spectrum policies at a software level rather than at a hardware level. And hypothetically, they would be more future proof, as spectrum opened up, the systems could be switched to take advantage of the new spectrum without hardware replacement.

This also could have some impact on RFID and sensor networks, especially as things move from region to region in which frequency and power specifications/requirements are different.

Kipp