Cross layer interference mitigation using a convergent two-stage game for ad hoc networks

In this paper, we concentrate on convergence issues of a two stage game namely on joint beamforming and channel allocation in an ad hoc network. Maximization of the received signal-to-interference-plus-noise (SINR) ratios between two communicating nodes under constant transmit power is considered. We propose a convergent sequential distortionless beamforming (SDRB) and cooperative channel allocation (CCA) algorithm which we call as SDRB_CCA, to mitigate the interference in the network. In this algorithm, a suboptimum beamforming algorithm is performed in PHY layer and a cooperative (or non-greedy) channel allocation is assigned based on the definition of potential games in MAC layer in an iterative manner. The payoff of each channel selection includes the interference received at each node as well as the interference that this selection will cause on the other nodes. Simulation results supported our analytical results in the sense that SDRB_CCA converges at the expense of decrease in average rate compared to the optimum joint iterative beamforming and greedy channel allocation (O_JIBGCA) algorithm especially in over-loaded scenarios.