Robust noncooperative rate-maximization game for MIMO Gaussian interference channels under bounded channel uncertainty

We propose a robust formulation for the noncooperative rate-maximization game in MIMO Gaussian interference channels under bounded channel uncertainty. The proposed robust game needs little additional computation and requires no additional information exchange among users when compared to the nominal game and thus maintains the low-complexity and distributed nature of the MIMO waterfilling algorithm. The robust rate-maximization game is shown to be equivalent to the nominal game with modified direct-channel matrices. The equilibrium solution of the robust rate-maximization game and the required iterative algorithm to obtain the solution are presented. Sufficient conditions for the uniqueness of the equilibrium and the convergence of the algorithm are also presented. Simulation results indicate that the robust solution in the presence of channel uncertainty performs better than the nominal solution with zero uncertainty, due to the users being more conservative in their power allocation when there is channel uncertainty.