In this paper, we explore the concept of integrated sensing and communication (ISAC) within a downlink cell-free massive MIMO (multiple-input multiple-output) system featuring multi-static sensing and users requiring ultra-reliable low-latency communications (URLLC). Our focus involves the formulation of two non-convex algorithms that jointly solve power and blocklength allocation for end-to-end (E2E) minimization. The objectives are to jointly minimize sensing/communication processing and transmission energy consumption, while simultaneously meeting the requirements for sensing and URLLC. To address the inherent non-convexity of these optimization problems, we utilize techniques such as the Feasible Point Pursuit - Successive Convex Approximation (FPP-SCA), Concave-Convex Programming (CCP), and fractional programming. We conduct a comparative analysis of the performance of these algorithms in ISAC scenarios and against a URLLC-only scenario where sensing is not integrated. Our numerical results highlight the superior performance of the E2E energy minimization algorithm, especially in scenarios without sensing capability. Additionally, our study underscores the increasing prominence of energy consumption associated with sensing processing tasks as the number of sensing receive access points rises. Furthermore, the results emphasize that a higher sensing signal-to-interference-plus-noise ratio threshold is associated with an escalation in E2E energy consumption, thereby narrowing the performance gap between the two proposed algorithms.

翻译：本文探讨了在下行无蜂窝大规模MIMO（多输入多输出）系统中集成感知与通信（ISAC）的概念，该系统具有多静态感知功能，且用户需要超可靠低延迟通信（URLLC）。我们的重点在于制定两种非凸算法，以联合求解用于端到端（E2E）最小化的功率和块长度分配。其目标是联合最小化感知/通信处理与传输能耗，同时满足感知和URLLC的要求。为解决这些优化问题固有的非凸性，我们采用了可行点追踪-逐次凸逼近（FPP-SCA）、凹凸规划（CCP）和分数规划等技术。我们对这些算法在ISAC场景下的性能进行了比较分析，并与未集成感知的纯URLLC场景进行了对比。数值结果表明，E2E能量最小化算法具有优越性能，尤其是在不具备感知能力的场景中。此外，我们的研究强调了随着感知接收接入点数量的增加，与感知处理任务相关的能耗日益显著。最后，结果强调，较高的感知信干噪比阈值会导致E2E能量消耗增加，从而缩小了两种算法之间的性能差距。