Comparative Study of SVD and QRS in Closed-Loop Beamforming Systems

TL;DR

This paper compares SVD and QRS algorithms for closed-loop beamforming, analyzing their advantages, limitations, and performance under different modulation and channel conditions.

Contribution

It provides a comparative analysis of SVD and QRS in closed-loop beamforming, highlighting conditions where each method performs better.

Findings

SVD enables parallelization of MIMO channels but has varying sub-channel gains.

QRS offers equal diagonal values but requires interference cancellation techniques.

Optimal modulation set selection can make SVD outperform QRS.

Abstract

We compare two closed-loop beamforming algorithms, one based on singular value decomposition (SVD) and the other based on equal diagonal QR decomposition (QRS). SVD has the advantage of parallelizing the MIMO channel, but each of the sub-channels has different gain. QRS has the advantage of having equal diagonal value for the decomposed channel, but the subchannels are not fully parallelized, hence requiring successive interference cancellation or other techniques to perform decoding. We consider a closed-loop system where the feedback information is a unitary beamforming matrix. Due to the discrete and limited modulation set, SVD may have inferior performance to QRS when no modulation set selection is performed. However, if the selection of modulation set is performed optimally, we show that SVD can outperform QRS.