On the distributed approaches with imperfect channel estimation for MIMO interference channel

Abstract

This paper considers distributed approaches for the downlink of multi-user MIMO interference network where all nodes iteratively and distributedly update their precoders or receive filters. Those include the distributed interference alignment and minimum sum mean-squared error schemes. We analyse these algorithms in presence of effective imperfect channel estimation at each iteration and we present the pre-coders and receive filters updates in this case. The numerical results show that the algorithms converge in average. However, the sum-rate performance is degraded with respect to the estimation errors. Based on these results, we show that for a useful design in practice it is preferable to ameliorate the quality of the estimators and reduce the number of iterations.