MMSE interference alignment algorithms with low bit rate feedback

Abstract

Interference alignment (IA) is a promising technique that allows high capacity gains in interfering channels. However to achieve the maximum degrees of freedom it requires the knowledge of the different link channel state information (CSI) between base station (BS) and user terminal (UT). A quantized version of the CSI is feedback from the UT to the BS, which is then used to perform the overall IA design. In this paper we propose several low-complexity and low-bit rate channel quantization techniques, requiring the quantization of only the non-zero components of the channel impulse response. The proposed techniques are evaluated under the minimum mean squared error (MMSE)-based IA algorithms for the downlink of OFDM signals. We show that to achieve performances close to the ones obtained with perfect channel knowledge, only a small number of bits is needed for channels quantization comparatively to the previous proposed techniques.