期刊: [Institute of Electrical and Electronics Engineers] 日期:2025-12-12卷期号:: 1832-1838
标识
DOI:10.1109/iccc68654.2025.11437700
摘要
With the rapid development of integrated sensing and communication (ISAC) networks and edge computing technologies, there is a growing demand for low-latency and highly reliable computing services among users. This paper investigates an integrated sensing, communication, and computing (ISCC) system for multi-unmanned aerial vehicle (UAV) swarms. The system is capable of sensing user requests within its coverage and providing computing and offloading services. To enhance the utilization of computational resources, we organize UAVs into computing groups with local collaboration capabilities and jointly optimize sensing accuracy and computational resource allocation. Aiming to minimize the total task processing latency of the UAV swarm, we design a distributed resource allocation algorithm based on federated and deep reinforcement learning for the ISCC system. Simulation results demonstrate that the proposed algorithm converges quickly. Moreover, compared to baseline algorithms, our approach significantly reduces processing latency, achieving a performance improvement of 26%.