Full-Duplex Enabled Integrating Communication and Computation: Joint Beamforming and Power Optimization With Imperfect CSI

This letter investigates a full-duplex enabled network that integrates communication and over-the-air computation (AirComp), where the base station (BS) performs uplink process for AirComp and downlink process for communication over the same time-frequency resources. Consider the case of imperfect channel state information, we formulate a stochastic sum rate maximization problem such that the mean square error of AirComp should be less than a given threshold and each node’s power budget is limited. To solve the stochastic optimization problem, we propose a fast stochastic algorithm which jointly optimizes the beamforming at the BS and the power factors of sensors, and the solution of each optimization variable has a closed or semi-closed form. The numerical results show the effectiveness of the proposed algorithm.