Software-defined joint routing and waveform selection for cognitive Ad Hoc networks

The problem of throughput maximization in a cognitive radio network with decentralized control (i.e., a cognitive radio ad hoc network) is considered in this paper. First, decentralized and localized algorithms for throughput maximization through joint cognitive routing and interference-avoiding waveform selection are proposed, based on a nonlinear optimization framework. The proposed algorithms adapt to time-varying traffic demands, interference profile, and network topology to locally maximize the achievable data rate while avoiding harmful interference to co-located primary or secondary users. Second, a prototype implementation of the proposed decentralized control algorithms is presented. The prototype is based on a software-defined-radio USRP2 platform that distributively senses and adapts its transmission based on results from real-time nonlinear optimization. Experiments on the software-defined-radio platform demonstrate the superior adaptivity of the proposed algorithm with respect to state-of-the-art non-cognitive approaches.