Superimposed Coding Based CSI Feedback Using 1-Bit Compressed Sensing

TL;DR

This paper introduces a novel 1-bit compressed sensing and superimposed coding method for CSI feedback in FDD massive MIMO systems, reducing bandwidth usage while improving feedback accuracy and error rates.

Contribution

The paper proposes a new superimposed coding scheme combined with 1-bit compressed sensing for efficient CSI feedback, enhancing accuracy without extra uplink bandwidth.

Findings

Improved bit error ratio for uplink user data

Enhanced accuracy of downlink CSI feedback

Reduced bandwidth consumption for CSI feedback

Abstract

In a frequency division duplex (FDD) massive multiple input multiple output (MIMO) system, the channel state information (CSI) feedback causes a significant bandwidth resource occupation. In order to save the uplink bandwidth resources, a 1-bit compressed sensing (CS)-based CSI feedback method assisted by superimposed coding (SC) is proposed. Using 1-bit CS and SC techniques, the compressed support-set information and downlink CSI (DL-CSI) are superimposed on the uplink user data sequence (UL-US) and fed back to base station (BS). Compared with the SC-based feedback, the analysis and simulation results show that the UL-US's bit error ratio (BER) and the DL-CSI's accuracy can be improved in the proposed method, without using the exclusive uplink bandwidth resources to feed DL-CSI back to BS.