A Reconfigurable Distributed Algorithm for K-User MIMO Interference Networks

It is already well-known that interference alignment (IA) achieves the sum capacity of the K-user interference channel at the high interference regime. On the other hand, it is intuitively clear that when the interference levels are very low, a sum-rate scaling of K (as opposed to K/2 for IA) should be accessed at high signal-to-noise ratio values by simple ("myopic") single-link multiple-input multiple-output (MIMO) techniques such as waterfilling. Recent results have indicated that in certain low-to-moderate interference cases, treating interference as noise may in fact be preferable. In this paper, we present a distributed iterative algorithm for K-user MIMO interference networks which attempts to adjust itself to the interference regime at hand, in the above sense, as well as to the channel conditions. The proposed algorithm combines the system-wide mean squared error minimization with the waterfilling solution to adjust to the interference levels and channel conditions and maximize accordingly each user's transmission rate. Sum-rate computer simulations for the proposed algorithm over Ricean fading channels show that, in the interference-limited regime, the proposed algorithm reconfigures itself in order to achieve the IA scaling whereas, in the low-to-moderate interference regime, it leads itself towards interference-myopic MIMO transmissions.