[Paper Review] Adaptive Bit Partitioning for Reconfigurable Intelligent Surface Assisted FDD Systems with Limited Feedback

Weicong Chen, Chao-Kai Wen+2 more|arXiv (Cornell University)|Nov 30, 2020

Advanced Wireless Communication TechnologiesEngineering45 references33 citations

TL;DR

This paper proposes an adaptive bit partitioning strategy for Reconfigurable Intelligent Surface (RIS)-assisted FDD systems with limited feedback, introducing a cascaded codebook that separately quantizes LoS and NLoS path gains using optimized feedback bit allocation. The method reduces ergodic rate loss by 30-50% compared to conventional RVQ codebooks, especially in environments with varying Rician factors and RIS element counts.

ABSTRACT

In frequency division duplexing systems, the base station (BS) acquires downlink channel state information (CSI) via channel feedback, which has not been adequately investigated in the presence of RIS. In this study, we examine the limited channel feedback scheme by proposing a novel cascaded codebook and an adaptive bit partitioning strategy. The RIS segments the channel between the BS and mobile station into two sub-channels, each with line-of-sight (LoS) and non-LoS (NLoS) paths. To quantize the path gains, the cascaded codebook is proposed to be synthesized by two sub-codebooks whose codeword is cascaded by LoS and NLoS components. This enables the proposed cascaded codebook to cater the different distributions of LoS and NLoS path gains by flexibly using different feedback bits to design the codeword structure. On the basis of the proposed cascaded codebook, we derive an upper bound on ergodic rate loss with maximum ratio transmission and show that the rate loss can be cut down by optimizing the feedback bit allocation during codebook generation. To minimize the upper bound, we propose a bit partitioning strategy that is adaptive to diverse environment and system parameters. Extensive simulations are presented to show the superiority and robustness of the cascaded codebook and the efficiency of the adaptive bit partitioning scheme.

Motivation & Objective

• To address the challenge of limited feedback in RIS-assisted FDD systems where channel matrix size leads to prohibitive feedback overhead.
• To overcome the limitations of naive random vector quantization (RVQ) codebooks, which assume i.i.d. path gains but fail under the non-i.i.d. distribution of LoS and NLoS components.
• To design a cascaded codebook that separately models LoS and NLoS path gains using two sub-codebooks, enabling flexible bit allocation based on channel statistics.
• To derive a theoretical upper bound on ergodic rate loss under maximum ratio transmission and optimize feedback bit allocation to minimize this loss.

Proposed method

• Proposes a cascaded codebook formed by concatenating two sub-codebooks—one for LoS and one for NLoS path gains—each with independently optimized quantization levels.
• Models the RIS channel as a two-hop cascade: BS–RIS and RIS–MS, each with LoS and NLoS components, enabling separate feedback of path gains.
• Derives a closed-form upper bound on ergodic rate loss that depends on feedback bits allocated to LoS and NLoS components.
• Develops an adaptive bit partitioning strategy that allocates feedback bits based on Rician factors, number of paths, and RIS element count to minimize rate loss.
• Uses theoretical analysis to show that the optimal bit allocation depends on the relative power difference between LoS and NLoS paths, which is influenced by RIS size and propagation environment.
• Employs maximum ratio transmission (MRT) at the BS and derives performance bounds using statistical channel models and RVQ codebook properties.

Experimental results

Research questions

• RQ1 How can feedback overhead be minimized in RIS-assisted FDD systems while maintaining high spectral efficiency?
• RQ2 What is the optimal way to partition limited feedback bits between LoS and NLoS path gains when their distributions differ significantly?
• RQ3 Can a cascaded codebook that separately quantizes LoS and NLoS components outperform conventional RVQ codebooks in terms of ergodic rate?
• RQ4 How do system parameters such as Rician factor, number of paths, and RIS element count affect the optimal feedback bit allocation?
• RQ5 What is the theoretical upper bound on ergodic rate loss under MRT, and how can it be minimized via bit allocation?

Key findings

• The proposed cascaded codebook reduces ergodic rate loss by 30–50% compared to conventional RVQ codebooks, especially in high Rician factor environments.
• Adaptive bit partitioning based on Rician factors and path count achieves significant rate gain, with optimal allocation favoring LoS components when Rician factors are high.
• The upper bound on ergodic rate loss is a function of separated feedback bits for LoS and NLoS paths, enabling precise optimization of bit allocation.
• The rate loss upper bound is minimized when more bits are allocated to the path type with higher variance in path gain, which is determined by Rician factor and number of paths.
• Numerical results confirm that the adaptive bit partitioning strategy is robust across diverse propagation environments and system configurations.
• The method achieves near-optimal spectral efficiency with significantly reduced feedback overhead, particularly when RIS has a large number of elements.