Advancing massive MIMO mm-Wave Channel Estimation by Coherence-Optimized Measurement Matrices

Articles in Press

Document Type : Research Article

Author

Assistant professor, Department of Electrical Engineering, IHCU, Theran, Iran

Abstract

In the realm of millimeter-wave (mmWave) communications, despite their promise of high data rates and expansive bandwidths, channel estimation encounters formidable challenges due to conspicuous path loss and the limited multipath components. This paper presents an innovative method that leverages the inherent sparsity of mmWave channels by operating within the two-dimensional transformed domain, this approach treats the channel as a sparse image representation. We advance the accuracy of sparse equivalent vectorized channel recovery by optimizing the measurement matrix. The proposed optimization method significantly reduces the requisite measurements and accelerates the estimation process and minimizes the mean squared error between the true and estimated channel matrices. Through comprehensive simulations, we evaluate our method against two scenarios: one where the compression rate is zero, and the sparse channel matrix recovery relies on the number of observations equating the number of channel matrix elements, and another where the compression rate is non-zero, but the measurement matrix remains unoptimized and randomly selected. Results demonstrate that our method outperforms the latter scenario and achieves accuracy comparable to the former, with significantly reduced computational overhead and accelerated computation speed.

Keywords

Main Subjects

AUT Journal of Electrical Engineering

Articles in Press, Accepted Manuscript
Available Online from 13 January 2025

Files

Share

How to cite

Statistics