Interference Alignment with Limited Feedback on Two-cell Interfering Two-User MIMO-MAC

TL;DR

This paper studies interference alignment in a two-cell two-user MIMO-MAC with limited feedback, proposing a feedback scheme that maintains multiplexing gain and analyzing the required feedback bits for optimal performance.

Contribution

It introduces a feedback framework for interference alignment under limited feedback and derives the scaling law for feedback bits needed to preserve multiplexing gain.

Findings

Feedback bits scale as $(M-1) imes ext{log}_2( ext{SNR}) + C$ to maintain multiplexing gain.

Sum-rate performance matches theoretical analysis in simulations.

Proposes a practical feedback scheme for interference alignment in multi-cell MIMO systems.

Abstract

In this paper, we consider a two-cell interfering two-user multiple-input multiple-output multiple access channel (MIMO-MAC) with limited feedback. We first investigate the multiplexing gain of such channel when users have perfect channel state information at transmitter (CSIT) by exploiting an interference alignment scheme. In addition, we propose a feedback framework for the interference alignment in the limited feedback system. On the basis of the proposed feedback framework, we analyze the rate gap loss and it is shown that in order to keep the same multiplexing gain with the case of perfect CSIT, the number of feedback bits per receiver scales as $B \geq (M\!-1\!)\!\log_{2}(\textsf{SNR})+C$, where $M$ and $C$ denote the number of transmit antennas and a constant, respectively. Throughout the simulation results, it is shown that the sum-rate performance coincides with the derived results.