Network-Assisted Full-Duplex RIS-Aided Cell-Free Massive MIMO for Integrated Sensing and Communication

Abstract

Integrated sensing and communication (ISAC) has emerged as a highly promising technology for future sixth-generation networks owing to its potential to optimize spectrum resources at minimal hardware costs. One promising realization of ISAC is the multi-static design, where several distributed nodes jointly enhance sensing and communication performance. However, the current multi-static designs fall short of maintaining accurate sensing performance while fulfilling the simultaneous uplink/downlink communication requirements. To solve this efficiently, we consider a reconfigurable intelligent surface (RIS) aided cell-free massive multiple-input multiple-output (CF-mMIMO) operating in network-assisted full-duplex (NAFD) mode and enhance the weighted sensing and communication rates of the network by efficiently designing transmit/receive beamformers at the APs and optimizing the RIS phase-shifts. Since the formulated problem is highly non-convex, we leverage tools such as quadratic-transform and succesive-convex approximation to convert the objective into a tractable form and then utilize Lagrange multipliers method to derive inter-AP-interference suppression enabled closed-form expressions for transmit and receive beamformers at the APs. Then, we propose a variance-reduced projected-gradient method, that relies on closed-form gradient updates to update the reflect beamformers at the RIS based on statistical-CSI. Simulation results demonstrate the effectiveness of the proposed design in achieving excellent ISAC performance at low computational complexity.