Structured Sensing Matrix Design for In-sector Compressed mmWave Channel Estimation

TL;DR

This paper introduces a structured sensing matrix design for in-sector compressed sensing in mmWave channel estimation, improving accuracy and reducing artifacts in the sector of interest.

Contribution

It proposes a novel class of structured CS matrices tailored for in-sector mmWave channel estimation, enhancing estimation quality over existing methods.

Findings

Lower aliasing artifacts in the sector of interest

Improved channel estimation accuracy

Effective in sub-Nyquist measurement regimes

Abstract

Fast millimeter wave (mmWave) channel estimation techniques based on compressed sensing (CS) suffer from low signal-to-noise ratio (SNR) in the channel measurements, due to the use of wide beams. To address this problem, we develop an in-sector CS-based mmWave channel estimation technique that focuses energy on a sector in the angle domain. Specifically, we construct a new class of structured CS matrices to estimate the channel within the sector of interest. To this end, we first determine an optimal sampling pattern when the number of measurements is equal to the sector dimension and then use its subsampled version in the sub-Nyquist regime. Our approach results in low aliasing artifacts in the sector of interest and better channel estimates than benchmark algorithms.