Talk Title:  Cooperative Strategy for Distributed Multi-point ISAC Systems

This work studies on ISAC systems for Mobile IoT, emphasizing distributed multi-point architectures that combine active and passive sensing. We first develop a power optimization framework for dual-function radar-communication systems, demonstrating that unlimited backhaul capacity with centralized signal fusion achieves higher target detection probability compared to limited backhaul relying on binary result fusion. Then we introduce a beamforming optimization strategy, distinguishing between the cases with and without sensing interference cancellation. It proves that dedicated sensing signals and optimal beam alignment significantly enhance sensing performance when interference is mitigated. We further propose a deep learning-based auto-compression framework using model-agnostic meta-learning (MAML) to address backhaul constraints. This approach efficiently compresses sensing signals, reducing overhead while maintaining detection accuracy, thus bridging ideal and limited backhaul scenarios. In summary, this work studies power and beamforming optimization issues in distributed multi-point ISAC architectures considering practical backhaul constraints, advancing spectral efficiency and sensing robustness for next-generation mobile IoT.