A rapidly adaptive AI is needed for maximum spectrum sharing and optimal data transfer within a network, possibly in the presence of hostile entities.

The goal of our Collaborative Intelligent Radio Networking (CIRN) research and development is to enable a rich spectral ecosystem to accommodate a wide variety of communicating devices while operating 100 to 1,000 times more efficiently than today’s wireless networks through a new paradigm of collaborative, local, and real-time decision making. In the military, there is growing reliance on unmanned platforms, from underwater sensors to satellites, and a push for broadband connectivity. However, there is also a growing shortage of RF spectrum. To maximize the RF spectrum, future radios will need to lose their isolation safety net and use greater intelligence to avoid interference. These radios will need to be able to collaborate directly with their peers to derive stable and satisfactory communications for all.

Designing a complete CIRN system solution that has a credible path for widespread adoption requires autonomous yet robust control and monitoring of the entire CIRN network. A practical and rapidly adaptive AI is needed to set several parameters for maximum spectrum sharing and optimal data transfer within a network, possibly in the presence of hostile entities. Each node must make decisions on its own as there is no guarantee of connectivity to a centralized node.

Our CIRN solutions seek the maximum spectrum sharing achievable across heterogeneous CIRN nodes by virtually eliminating the most common overheads associated with existing dynamic spectrum sharing approaches such as channel rendezvous, collision avoidance, quasi-static frequency division based sharing, separation of environment sensing and communication functions, and retransmission-based error correction.