Dynamic Quantization based IA for Homogeneous and three-tier Heterogeneous Cellular Systems

Abstract

Interference is believed to be the most significant difficulty in next-generation wireless networks, where it limits the achievable throughput. Interference alignment (IA) is considered a satisfying solution that has the capability of completely canceling the interference and hence, enhancing the achievable data rate. In this paper, we propose a dynamic quantization-based IA scheme that uses an iterative IA approach, namely the rank constrained rank minimization (RCRM) IA, which aims to maximize the number of achievable degrees-of-freedom (DoF) through minimizing the rank of interference matrices. To the best of our knowledge, most of the previous works that are using RCRM-IA have focused on networks with perfect channel state information at the transmitters and symmetrical interference topology. Therefore, in this paper, we implement RCRM-IA in a limited feedback system with path loss and spatial correlations at transmitter side. Moreover, we conduct our experiments using two different networks or system models; 1) a symmetric homogeneous network with fixed parameters and 2) an asymmetric heterogeneous network (RetNet). Results show that the dynamic quantization method enhances the achievable rate compared to conventional quantization methods.