Interference Alignment Assisted by D2D Communication for the Downlink of MIMO Heterogeneous Networks

Abstract

To cope with the conflict between the increasing data traffic demands and scarcity of available spectrum resources in wireless networks, multi-tier heterogeneous networks within the same spectrum have been considered to be deployed to provide higher network capacity. However, the spectrum-coexist leads to serious interference. This paper investigates a novel interference alignment (IA) scheme assisted by device-to-device (D2D) communication (DaIA) under a multiple-input multiple-output Gaussian interference broadcast channel heterogeneous scenario. We focus on a hotspot area of communication within the macro cell coverage in which there are two small cells. There is already existing IA algorithm for the case that macro cell base station (BS) has two users. In this paper, we consider different setting that macro cell BS can serve more users. First, as an important aspect of IA, the feasibility conditions for the considered scenario are presented. It acts as a guideline to design the IA scheme. Then the DaIA scheme is proposed to efficiently manage the co-tier interference and inter-tier interference with the reduced channel state information (CSI). Specifically, the proposed DaIA scheme jointly designs precoders and receive filters in closed-form in which the paired users can directly interact with the CSI of interfering channels to design their receiving filters by exploiting D2D communication. Thus the total amount of effective CSI feedback acquired at the BS is greatly reduced. Moreover, we analyze the achievable degrees of freedom of the proposed DaIA scheme. Finally, for the small cells, the precoders and receive filters are further optimized to maximize their achievable rate by combining the water-filling algorithm and the singular value decomposition of the equivalent channel matrix. The simulation results indicate that the proposed DaIA scheme outperforms the conventional interference mitigation schemes.