Cyclic interference alignment for MIMO interference channels: A hybrid approach of MTLI and PSO

The paper presents an interference alignment (IA) scheme for wireless systems with multiple-input-multiple-output (MIMO) interference channels. The scheme executes the sum rate optimization task in an iterative cycle. Specifically, in each cycle, the minimum total leakage interference (MTLI) method is used initially to find a tentative solution candidate. Next, the particle swarm optimization (PSO) method uses the tentative candidate to assist the joint search for feasible solutions that will maximize the total sum rate. Then, at the end of each cycle, the best solution candidate derived by PSO is passed to the next cycle to help the MTLI mechanism refine the candidate. The cyclic cooperation among the mechanisms of MTLI's receivers, MTLI's transmitters, and PSO allows the candidates to be refined cycle by cycle. The results of simulations and complexity analysis show that the proposed IA scheme not only reduces the computational complexity significantly, but also achieves a better performance gain than four existing approaches.