联邦学习驱动的低空通感节点自适应切换

Federated learning-driven adaptive switching of low-altitude ISAC nodes

摘要

为应对复杂电磁环境下的感知连续性及高精度挑战，并充分利用现有通信基础设施，提出一种联邦学习驱动的低空通感一体化网络感知节点自适应切换机制。首先，采用层级化的混频基站部署与基于收发分离的多发单收感知网络架构，构建协同感知网络模型；其次，针对高空目标空域位置的强动态特性，利用长短期记忆网络预测其运动轨迹；进而引入联邦学习框架，在分布式节点间协同训练一个全局的“节点调度决策模型”，并以环境状态与目标预测信息为输入，实时输出最优的收发节点组合策略；最后，依托动态选举的中央节点，实现感知节点的快速自适应切换，动态匹配业务需求并持续跟踪目标。仿真结果表明，在高速机动目标全程跟踪过程中，该算法在感知覆盖率与资源利用效率等方面均优于传统感知方案，验证了该方法的有效性与优越性。

Abstract

With the continuous evolution of integrated sensing and communications (ISAC) technology and the rapid development of the low-altitude economy, there is an increasingly urgent demand for continuous and accurate sensing of highly maneuverable targets in low-altitude heterogeneous networks. To address the challenges of sensing continuity and high precision in complex electromagnetic environments, this paper proposes a federated learning-driven adaptive handover mechanism for sensing nodes in low-altitude ISAC networks by leveraging existing communication infrastructure. First, a hierarchical hybrid-frequency base station deployment and a multi-transmit single-receive sensing architecture based on separate transmission and reception are employed to construct the foundation for cooperative sensing. Second, in view of the highly dynamic spatial trajectories of aerial targets, a long short-term memory network is utilized to predict target motion. Furthermore, a federated learning framework is introduced to collaboratively train a global “node scheduling decision model” among distributed nodes. This model takes environmental states and target prediction information as input and outputs an optimal transceiver node combination strategy in real time. Finally, through a dynamically elected central node, rapid adaptive handover of sensing nodes is achieved, dynamically aligning with service requirements and enabling persistent target tracking. Simulation results demonstrate that, during full-course tracking of high-speed maneuvering targets, the proposed algorithm significantly outperforms conventional sensing schemes in terms of perception coverage rate and resource utilization efficiency, verifying the effectiveness and superiority of the proposed approach.

关键词

低空通感一体化网络 / 联邦学习 / 收发分离 / 感知节点切换

Key words

low-altitude integrated sensing and communications network / federated learning / separate transmission and reception / sensing node switching

参考文献

[1] 巨涛, 李林娟, 张文金, 等. 多无人机辅助的移动边缘计算任务卸载及路径优化方法[J].电子科技大学学报, 2025, 54(1): 72-83.

[2] Ju Tao, Li Linjuan, Zhang Wenjin, et al. MATOPO: A multi-UAV assisted task offloading and path optimization method for moving edge computing[J].Journal of University of Electronic Science and Technology of China, 2025, 54(1): 72-83. (in Chinese)

[3] 李新民, 张徐浩, 刘易阳, 等. 基于强化学习的无人机通信感知一体化能耗优化[J].电子科技大学学报, 2025, 54(1): 23-28,51.

[4] Li Xinmin, Zhang Xuhao, Liu Yiyang, et al. Energy optimization algorithm for ISAC-enabled unmanned aerial vehicles system via reinforcement learning[J].Journal of University of Electronic Science and Technology of China, 2025, 54(1): 23-28,51. (in Chinese)

[5] Páez-Quinde C, Arroba-Freire E, Espinosa-Jaramillo M T, et al. Gamification as collaborative learning resources in technological education[C]//Proceedings of 2023 IEEE Global Engineering Education Conference. Kuwait: IEEE, 2023: 1-5.

[6] Xue Zheng, Liu Chang, Wen Fuxi, et al. Joint optimization of task offloading and resource allocation for cooperative perception in vehicular edge computing systems[J].IEEE Transactions on Vehicular Technology, 2025, 74(12): 19237-19251.

[7] D'Amato A, Dantas M, De Macedo D D J. MAROQ: A resource allocation model driven through quality of experience[J].International Journal of Grid and Utility Computing, 2017, 8(2): 82-93.

[8] Wang Jianwei, Chen Shuo. Deep reinforcement learning-based secrecy rate optimization for simultaneously transmitting and reflecting reconfigurable intelligent surface-assisted unmanned aerial vehicle-integrated sensing and communication systems[J].Sensors, 2025, 25(5): 1541.

[9] Zhan Yan, Hong Yi, Li Deying, et al. Joint optimization of DCCR and energy efficiency in active STAR-RIS-assisted UAV-NOMA networks[J].Drones, 2025, 9(8): 520.

[10] Megahed A, El-Haleem A M A, Elmesalawy M M, et al. Deep learning optimization of STAR-RIS for enhanced data rate and energy efficiency in 6G wireless networks[J].Scientific Reports, 2025, 15(1): 26311.

[11] Lin Hongyi, Liu Yang, Wang Liang, et al. Big data-driven advancements and future directions in vehicle perception technologies: From autonomous driving to modular buses[J].IEEE Transactions on Big Data, 2025, 11(3): 1568-1587.

[12] Li Zongdian, Wang Kui, Yu Tao, et al. Het-SDVN: SDN-based radio resource management of heterogeneous V2X for cooperative perception[J].IEEE Access, 2023, 11: 76255-76268.

[13] Dai Bin, Xu Fanglin, Cao Yuanyuan, et al. Hybrid sensing data fusion of cooperative perception for autonomous driving with augmented vehicular reality[J].IEEE Systems Journal, 2021, 15(1): 1413-1422.

[14] Zaki A M, Elsayed S A, Elgazzar K, et al. Quality and budget-oriented task offloading for vehicular cooperative perception using reinforcement learning[J].IEEE Internet of Things Journal, 2025, 12(14): 26169-26185.

[15] Tagliaferri D, Manzoni M, Mizmizi M, et al. Cooperative coherent multistatic imaging and phase synchronization in networked sensing[J].IEEE Journal on Selected Areas in Communications, 2024, 42(10): 2905-2921.

[16] Kim S W, Qin Baoxing, Chong Zhuangjie, et al. Multivehicle cooperative driving using cooperative perception: Design and experimental validation[J].IEEE Transactions on Intelligent Transportation Systems, 2015, 16(2): 663-680.

[17] Alkharabsheh E, Alawneh S, Rawashdeh O. A cloud-based cooperative perception system for object detection using roadside cameras[C]//2025 IEEE 4th International Conference on Computing and Machine Intelligence. Mount Pleasant: IEEE, 2025: 1-5.

[18] Rauch A, Klanner F, Dietmayer K. Analysis of V2X communication parameters for the development of a fusion architecture for cooperative perception systems[C]//2011 IEEE Intelligent Vehicles Symposium. Baden-Baden: IEEE, 2011: 685-690.

基金资助

南京市重大科技专项(202309010)

金陵科技学院高层次人才科研启动项目(jit-b-201632)

国家级大学生创新创业训练计划项目(202513573036X)