Interplay between Sensing and Communication in Cell-Free Massive MIMO with URLLC Users
Zinat Behdad, Özlem Tuğfe Demir, Ki Won Sung, Cicek Cavdar
TL;DR
This work tackles ISAC in the downlink of a cell-free massive MIMO system serving URLLC users while performing multi-static sensing. It introduces SeURLLC+, a Feasible Point Pursuit - Successive Convex Approximation (FPP-SCA) based power-allocation algorithm that maximizes energy efficiency under both sensing SINR and URLLC reliability constraints, and it defines a network availability metric capturing joint ISAC-URLLC QoS. The approach reformulates the problem with SOC constraints and handles non-convex sensing requirements via iterative convexification and slack variables, yielding a feasible solution that balances sensing and communication needs. Numerical results show that URLLC-only designs cannot satisfy sensing requirements, while SeURLLC+ achieves both with additional sensing signaling; however, energy penalties exist, which can be mitigated by dedicating more symbols to sensing and by increasing blocklength to improve robustness.
Abstract
This paper studies integrated sensing and communication (ISAC) in the downlink of a cell-free massive multiple-input multiple-output (MIMO) system with multi-static sensing and ultra-reliable low-latency communication (URLLC) users. We propose a successive convex approximation-based power allocation algorithm that maximizes energy efficiency while satisfying the sensing and URLLC requirements. In addition, we provide a new definition for network availability, which accounts for both sensing and URLLC requirements. The impact of blocklength, sensing requirement, and required reliability as a function of decoding error probability on network availability and energy efficiency is investigated. The proposed power allocation algorithm is compared to a communication-centric approach where only the URLLC requirement is considered. It is shown that the URLLC-only approach is incapable of meeting sensing requirements, while the proposed ISAC algorithm fulfills both sensing and URLLC requirements, albeit with an associated increase in energy consumption. This increment can be reduced up to 75% by utilizing additional symbols for sensing. It is also demonstrated that larger blocklengths enhance network availability and offer greater robustness against stringent reliability requirements.