DARPA's Latest Grand Challenge Takes On The Radio Spectrum (gizmag.com)

An anonymous reader quotes a report from Gizmag: One of the most hotly contested bits of real estate today is one you can't see. As we move into an increasingly wireless-connected world, staking out a piece of the crowded electromagnetic spectrum becomes more important. DARPA is hoping to help solve this issue with its latest Grand Challenge, which calls for the use of machine-learning technologies to enable devices to share bandwidth. The Spectrum Collaboration Challenge (SC2) is based on the idea that wireless devices would work better if they cooperated with one another rather than fought for bandwidth. Since not all devices are active at all times, the agency says, it should be possible through the use of artificial intelligence machine-learning algorithms to allow them to figure out how to share the spectrum with a minimum of conflict. DARPA announced the competition in front of 8000 engineers on Wednesday at the International Wireless Communications Expo (IWCE) in Las Vegas. SC2 will run from 2017 through 2020 with teams competing to create radios that can collaborate most effectively with other radios. The competition will end with a live event and the prize is $2 million.

Comments

Re:10 pounds of shit...

[plover]

They could do what cellular radios do to increase sharing: cut transmit power until both radios agree the message is acceptably low but still reliably receivable. Then, when you need to increase the density of transmitters, you do so by distributing more access points. It also helps conserve transmit power, which is good for battery life.

And if they don't think CSMA/CD is effective enough, once the density of collisions is too high, they could let the access points switch to a time slice mode. They could be either self clocking, or timed on a separate shared control frequency.

Re:That's great... for the USA

[sociocapitalist]

But the rest of the world and considering illegal frequencies being used around the US will make this problem harder than one thinks.

So you come up with a great solution for 900Mhz, problem is you can't use it in the EU or Asia. Or how about 433? nope, more conflicts outside the US.

Then there's the 1W vs 25mW (rest of the world) requirement....

Not only is this a tech and regulation problem, it's a frequency management and standards nightmare.

At least you can drive a DARPA challenge car or a biped robot in pretty much any country without conflicting tech (but conflicting regs).

They're looking for a mechanism to share bandwidth intelligently. There is nothing specific to frequencies or power handling in how bandwidth might be shared so the result should be usable regardless of local variations on implementation.