[Paper Review] Downlink Pilot Precoding and Compressed Channel Feedback for FDD-Based Cell-Free Systems

Seungnyun Kim, Jun Won Choi+1 more|arXiv (Cornell University)|Sep 16, 2019

Advanced MIMO Systems Optimization17 references4 citations

TL;DR

This paper proposes a downlink pilot precoding and compressed channel feedback scheme for FDD-based cell-free massive MIMO systems to reduce CSI feedback overhead. By exploiting angle reciprocity to estimate dominant angle of departure (AoD) from uplink pilots and feeding back only path gain information (PGI) of the strongest paths, the scheme achieves over 80% feedback overhead reduction compared to conventional CSI feedback, with minimal rate degradation due to reduced quantization distortion.

ABSTRACT

Cell-free system where a group of base stations (BSs) cooperatively serves users has received much attention as a promising technology for the future wireless systems. In order to maximize the cooperation gain in the cell-free systems, acquisition of downlink channel state information (CSI) at the BSs is crucial. While this task is relatively easy for the time division duplexing (TDD) systems due to the channel reciprocity, it is not easy for the frequency division duplexing (FDD) systems due to the CSI feedback overhead. This issue is even more pronounced in the cell-free systems since the user needs to feed back the CSIs of multiple BSs. In this paper, we propose a novel feedback reduction technique for the FDD-based cell-free systems. Key feature of the proposed technique is to choose a few dominating paths and then feed back the path gain information (PGI) of the chosen paths. By exploiting the property that the angles of departure (AoDs) are quite similar in the uplink and downlink channels (this property is referred to as angle reciprocity), the BSs obtain the AoDs directly from the uplink pilot signal. From the extensive simulations, we observe that the proposed technique can achieve more than 80% of feedback overhead reduction over the conventional CSI feedback scheme.

Motivation & Objective

• Address the high CSI feedback overhead in FDD-based cell-free massive MIMO systems, where users must report CSI for multiple distributed base stations.
• Overcome the limitations of TDD-based cell-free systems, such as uplink-downlink switching delays and RF calibration errors, by enabling practical FDD operation.
• Reduce feedback overhead without sacrificing spectral efficiency by exploiting channel sparsity and angle reciprocity in the spatial domain.
• Design a feedback-compressed pilot precoding scheme that enables efficient downlink beamforming with minimal feedback bits.
• Demonstrate that feedback overhead scales linearly with the number of dominant paths, not the number of transmit antennas, enabling scalable deployment.

Proposed method

• Use spatially precoded downlink pilots to enable the user to estimate the dominant path components (AoD and PGI) in the downlink.
• Leverage angle reciprocity to estimate the angle of departure (AoD) at the base stations directly from uplink pilot signals, eliminating the need to feed back AoD.
• Feed back only the path gain information (PGI) of the K strongest paths, selected to maximize sum rate, reducing feedback bits from O(M) to O(K), where K << M.
• Apply a compressed feedback framework where the user quantizes and reports only the PGI of dominant paths, while the base stations reconstruct the effective channel using estimated AoDs.
• Use a structured channel model where the channel is represented as a sum of K dominant multipath components, each characterized by AoD and PGI.
• Formulate the effective SINR at the user as a function of the largest eigenvalue of a structured matrix, which is analytically shown to be L+1, enabling performance analysis.

Experimental results

Research questions

• RQ1 Can angle reciprocity be effectively exploited to eliminate the need to feed back AoD information in FDD-based cell-free systems?
• RQ2 To what extent can feedback overhead be reduced by reporting only path gain information (PGI) of dominant multipath components?
• RQ3 How does the performance of the proposed PGI feedback scheme compare to conventional CSI feedback in terms of spectral efficiency and feedback overhead?
• RQ4 Does the feedback overhead scale with the number of dominant paths rather than the number of transmit antennas?
• RQ5 What is the theoretical performance limit of the proposed scheme in terms of achievable rate, and how close is it to the ideal case with perfect PGI?

Key findings

• The proposed PGI feedback scheme achieves over 80% reduction in feedback overhead compared to conventional CSI feedback schemes in realistic FDD-based cell-free systems.
• The feedback overhead scales linearly with the number of dominant paths (K), not the number of transmit antennas (M), enabling scalable operation in massive MIMO settings.
• The system with compressed PGI feedback achieves a sum rate within a constant gap of the ideal system with perfect PGI, due to low quantization distortion.
• The largest eigenvalue of the effective channel matrix is analytically proven to be L+1, where L is the number of dominant paths, enabling tight rate analysis.
• The performance gain of the proposed scheme increases with the number of cooperating base stations, highlighting its scalability in dense networks.
• The scheme maintains high spectral efficiency even with reduced feedback, as the quantization distortion is significantly lower than in conventional CSI feedback.