Feedback Constrained Interference Alignment Enabled by PCA Codebook Design for 6G Era

Interference Alignment (IA) has great potential to effectively reduce interference in multi-cell multi-user networks at high signal-to-noise (SNR) ratios in future 6G. However, the implementation of IA depends on the perfect global channel state information (CSI) at all transceivers. A large number of CSI feedback to the transmitter will greatly occupy the uplink channel bandwidth. Thus, IA with limited feedback is introduced to reduce the feedback bits. In this paper, in order to suppress the interference and reduce the amount of feedback in a multi-cell multi-user MIMO network, a novel interference alignment codebook model design along with a low complexity and decentralized scheme is introduced. The distance between sample points of this model is analyzed, constructing our codebook model by mapping all channel state information to a hyperplane with the goal of maximizing the distance between sample points on the plane. Considering that each receiver is subject to interference from different transmitters, we design the precoder matrix with the aim of optimizing the total distance of interference at all receivers. The simulation results show that the proposed scheme can achieve better performance than the traditional scheme.