Robust Lattice Alignment for K-user MIMO Interference Channels with Imperfect Channel Knowledge

TL;DR

This paper introduces a robust lattice alignment technique for K-user MIMO interference channels that effectively manages imperfect channel knowledge across all SNR levels, improving data rates and decoding reliability.

Contribution

It proposes a novel robust lattice alignment method with a two-stage decoding algorithm tailored for quasi-static MIMO channels with imperfect CSI, addressing limitations of existing methods.

Findings

Achieves higher data rates under imperfect CSI.

Demonstrates robustness across all SNR regimes.

Reduces complexity with an iterative optimization algorithm.

Abstract

In this paper, we consider a robust lattice alignment design for K-user quasi-static MIMO interference channels with imperfect channel knowledge. With random Gaussian inputs, the conventional interference alignment (IA) method has the feasibility problem when the channel is quasi-static. On the other hand, structured lattices can create structured interference as opposed to the random interference caused by random Gaussian symbols. The structured interference space can be exploited to transmit the desired signals over the gaps. However, the existing alignment methods on the lattice codes for quasi-static channels either require infinite SNR or symmetric interference channel coefficients. Furthermore, perfect channel state information (CSI) is required for these alignment methods, which is difficult to achieve in practice. In this paper, we propose a robust lattice alignment method for quasi-static MIMO interference channels with imperfect CSI at all SNR regimes, and a two-stage decoding algorithm to decode the desired signal from the structured interference space. We derive the achievable data rate based on the proposed robust lattice alignment method, where the design of the precoders, decorrelators, scaling coefficients and interference quantization coefficients is jointly formulated as a mixed integer and continuous optimization problem. The effect of imperfect CSI is also accommodated in the optimization formulation, and hence the derived solution is robust to imperfect CSI. We also design a low complex iterative optimization algorithm for our robust lattice alignment method by using the existing iterative IA algorithm that was designed for the conventional IA method. Numerical results verify the advantages of the proposed robust lattice alignment method.