Atomic Norm Denoising-Based Joint Channel Estimation and Faulty Antenna Detection for Massive MIMO

TL;DR

This paper introduces a novel method combining atomic norm denoising with ADMM to jointly estimate channels and detect faulty antennas in massive MIMO systems, improving accuracy and efficiency.

Contribution

It formulates joint channel estimation and faulty antenna detection as an extended atomic norm denoising problem and proposes an efficient ADMM-based solution.

Findings

The proposed method effectively detects faulty antennas.

It outperforms existing approaches in accuracy.

The fast algorithm approximates the atomic norm minimization well.

Abstract

We consider joint channel estimation and faulty antenna detection for massive multiple-input multiple-output (MIMO) systems operating in time-division duplexing (TDD) mode. For systems with faulty antennas, we show that the impact of faulty antennas on uplink (UL) data transmission does not vanish even with unlimited number of antennas. However, the signal detection performance can be improved with a priori knowledge on the indices of faulty antennas. This motivates us to propose the approach for simultaneous channel estimation and faulty antenna detection. By exploiting the fact that the degrees of freedom of the physical channel matrix are smaller than the number of free parameters, the channel estimation and faulty antenna detection can be formulated as an extended atomic norm denoising problem and solved efficiently via the alternating direction method of multipliers (ADMM). Furthermore, we improve the computational efficiency by proposing a fast algorithm and show that it is a good approximation of the corresponding extended atomic norm minimization method. Numerical simulations are provided to compare the performances of the proposed algorithms with several existing approaches and demonstrate the performance gains of detecting the indices of faulty antennas.