Joint Precoding and AP Selection for Energy Efficient RIS-aided Cell-Free Massive MIMO Using Multi-agent Reinforcement Learning

TL;DR

This paper investigates the joint precoding and access point (AP) selection for energy efficient RIS-aided CF mMIMO system and proposes an adaptive power threshold-based AP selection scheme to further enhance the EE of the considered system.

Abstract

Cell-free (CF) massive multiple-input multiple-output (mMIMO) and reconfigurable intelligent surface (RIS) are two advanced transceiver technologies for realizing future sixth-generation (6G) networks. In this paper, we investigate the joint precoding and access point (AP) selection for energy efficient RIS-aided CF mMIMO system. To address the associated computational complexity and communication power consumption, we advocate for user-centric dynamic networks in which each user is served by a subset of APs rather than by all of them. Based on the user-centric network, we formulate a joint precoding and AP selection problem to maximize the energy efficiency (EE) of the considered system. To solve this complex nonconvex problem, we propose an innovative double-layer multi-agent reinforcement learning (MARL)-based scheme. Moreover, we propose an adaptive power threshold-based AP selection scheme to further enhance the EE of the considered system. To reduce the computational complexity of the RIS-aided CF mMIMO system, we introduce a fuzzy logic (FL) strategy into the MARL scheme to accelerate convergence. The simulation results show that the proposed FL-based MARL cooperative architecture effectively improves EE performance, offering a 85\% enhancement over the zero-forcing (ZF) method, and achieves faster convergence speed compared with MARL. It is important to note that increasing the transmission power of the APs or the number of RIS elements can effectively enhance the spectral efficiency (SE) performance, which also leads to an increase in power consumption, resulting in a non-trivial trade-off between the quality of service and EE performance.

Figures (12)

• Figure 1: Illustration of two representative RIS-aided CF mMIMO systems: (a) AP full coverage for all UEs; (b) User-centric: AP selects a portion of UEs to serve.
• Figure 2: Illustration of a double-layer FL-based MARL network, which incorporates the joint AP selection and precoding network and the RIS beamforming network. The first layer network jointly designs the precoding matrix of all APs and adopts an adaptive power threshold-based algorithm for AP selection; The second layer network designs the RIS beamforming based on the output results of the first layer network.
• Figure 3: Convergence rate of MADDGP for AP selection or AP full coverage scenarios ($L = 8$, $M = 2$, $K = 6$, $N = 64$, $P_{\rm{AP,max}} = 5\,\rm{dB}$, $P_{\rm{element}} = -20 \,\,\rm{dB}$).
• Figure 4: Convergence rate of FL-based MADDGP for AP selection or AP full coverage scenarios ($L = 8$, $M = 2$, $K = 6$, $N = 64$, $P_{\rm{AP,max}} = 5\,\rm{dB}$, $P_{\rm{element}} = -20\,\,\rm{dB}$).
• Figure 5: Convergence time of MADDPG and FL-MADDPG algorithm with AP selection ($L = 8$, $M = 2$, $K = 6$, $N = 64$, $P_{\rm{AP,max}} = 5\,\,\rm{dB}$, $P_{\rm{element}} = -20\,\,\rm{dB}$).

Theorems & Definitions (2)

• Remark 1
• Remark 2