A game theoretical approach for coded cooperation in cognitive radio networks

Stupia, Ivan [UCL] Vandendorpe, Luc [UCL] Andreotti, R. Lottici, V.

In this paper, the authors focus on a game theoretical approach for cooperation in cognitive radio (CR) networks. In particular, an integrated design framework which relies on the effective SNR methodology is proposed. The authors derive a distributed power allocation policy aimed at maximizing the reliability of a cooperative BIC OFDM link wherein pragmatic modulation and coding schemes are considered. More in detail, the cognitive devices adapt their power to enable efficient cooperation and coexistence between cognitive nodes and primary networks. First of all, the gain due to the cooperation protocol is analytically derived, resorting to a simple first order recursive equation that depends on the current channel conditions. Then, after an accurate formalization of the optimization problem, a distributed iterative solution based on a novel algorithm, named Successive Set Reduction, is proposed. In particular, the authors show that : i) the proposed power allocation policy takes into account the cooperative gain through a simple scalar value, named cooperative effective SNR; ii) it is effective in improving the packet error rate performance with respect to other conventional power allocation strategies, thus allowing a better coverage for the secondary network; iii) the convergence of the distributed algorithm has exponential speed and requires only local signaling between secondary users. © 2012 IEEE.