[Battlemesh] DARPA and The Spectrum Collaboration Challenge

[Mitar]

Hi!

(Note: This challenge is funded by US military.)

"The DARPA Spectrum Collaboration Challenge (SC2) is the
first-of-its-kind collaborative machine-learning competition to overcome
scarcity in the radio frequency (RF) spectrum. Today, spectrum is
managed by dividing it into rigid, exclusively licensed bands. This
human-driven process is not adaptive to the dynamics of supply and
demand, and thus cannot exploit the full potential capacity of the
spectrum. In SC2, competitors will reimagine a new, more efficient
wireless paradigm in which radio networks autonomously collaborate to
dynamically determine how the spectrum should be used moment to moment.

The team whose radio design most reliably achieves successful
communication in the presence of other competing radios could win as
much as $3,500,000."

https://spectrumcollaborationchallenge.com/

Interesting it is that they are doing their own "battle" first, to
create a dataset:

"DARPA is organizing a Battle of the ModRecs Workshop, March 6-8 at
DySPAN 2017.

The Battle of the ModRecs seeks to explore the field of modulation
recognition to better understand how to approach these problems and,
most importantly, to compare results. Data driven approaches, which have
seen exponential adoption in other AI community problems, are beginning
to emerge in the spectrum community.  The goal of the workshop is to
spur the community to begin building and using commonly available
datasets. These datasets can serve as benchmarks for open comparisons in
much the same way that ImageNet and MNIST serve as benchmarks for the
image recognition community.

The Battle is separated into a workshop and a live over-the-air event.
The workshop will take place during the DySPAN workshop day of March 6,
2017. The Battle will take place between March 7 and 8, 2017 during the
technical conference."

"The Battle of the ModRecs"? I think somebody is copying us.


Mitar

-- 
http://mitar.tnode.com/
https://twitter.com/mitar_m