Federated Learning Based Integrated Sensing, Communications, and Powering Over 6G Massive-MIMO Mobile Networks

To sense mobile users' positions and channel states while simultaneously conveying information for supporting the 6G traffic, the integrated sensing and communication (ISAC) has been proposed as a key technique for sensing and communicating using the same radio frequency and hardware. Moreover, to sup-port energy-constrained and battery-limited wireless networks, simultaneous wireless information and power transfer (SWIPT) has also emerged to simultaneously deliver information and energy to a receiver. To integrate these two techniques, integrated sensing, communications, and powering (ISACP) has been developed to simultaneously sense the targeted mobile users, transmit the information, and deliver the power. However, how to efficiently optimize the power allocation in ISACP schemes to support the 6G traffic has imposed many new challenges not encountered before. To overcome these challenges, in this paper we propose a federated-learning (FL) enabled ISACP scheme to maximize the energy-efficiency over 6G massive MIMO wireless networks. First, we establish the system models for our proposed ISACP and FL schemes. Second, we design the training data set collection in the FL algorithm by sending the pilot signal from mobile users to the massive MIMO base station, and formulate an energy-efficiency maximization problem for the developed ISACP scheme. Third, To solve the formulated energy-efficiency maxi-mization problem, we propose an actorcritic enabled multi-agent based FL mechanism. Finally, we use the numerical analyses to validate and evaluate our developed schemes.