Joint detection and localization of moving targets in multi-cell ISAC systems

In this paper, we investigate the problem of target localization in multi-cell integrated sensing and communication (ISAC) systems and propose a joint detection and localization method for moving targets. We first develop a comprehensive transceiver signal model for multi-cell ISAC systems and design a corresponding precoding scheme to ensure effective coordination between sensing and communication tasks. Unlike conventional methods that separate detection and localization, our approach unifies them within a single model. By directly processing the received baseband signals from all nodes, we jointly estimate the target positions in the state space while applying a detection threshold to ensure reliability, thereby avoiding the performance degradation caused by decoupled processing. We further propose two tailored solutions for different network scenarios: one for fully networked multi-cell ISAC systems, suitable for systems with high-bandwidth and stable connections, and another for partially networked multi-cell ISAC systems, applicable to bandwidth-constrained or dynamically connected systems. For the latter case, where the joint detection and localization problem becomes a high-dimensional non-convex optimization with unknown parameters, we develop an efficient iterative solution based on the Expectation-Maximization (EM) algorithm. In addition, we derive the Cramér-Rao lower bound (CRLB) of the proposed method, establishing a theoretical benchmark for its estimation performance. Numerical simulations demonstrate the effectiveness of the proposed method in achieving reliable target detection and localization in ISAC systems.