A new opportunistic interference alignment scheme and performance comparison of MIMO interference alignment with limited feedback

We investigate the performance of several interference alignment (IA) schemes for a three-user MIMO channel with limited feedback. In our study, we consider traditional IA with quantized channel state information at each transmitter, and opportunistic IA (OIA) that exploits the idea of opportunistic selection to select users whose interfering channels are most aligned. The performance of the former is determined by the quantization resolution, whereas the performance of the later depends on the number of users available for selection. We show that OIA only requires a small number of users to achieve a comparable performance to traditional IA schemes with quantized feedback. Motivated by this fact, we propose a new user selection criterion that outperforms the existing state of the art. Simulations show the superior performance of our OIA scheme under different strong/mild interference conditions.