Optimized design for integrated sensing and communication in secure MIMO SWIPT systems

Abstract

This paper investigates a secure multiple-input multiple-output integrated sensing and communication (MIMO ISAC)-enabled untrusted simultaneous wireless information and power transfer (SWIPT) system in which a base station is concurrently responsible for three different functionalities of communicating with users, sensing targets, and powering energy harvesting (EH) devices. These EH devices are assumed to be untrusted and potentially wiretap the information intended for legitimate users. This paper considers three different system design strategies: (i) communication-centric design, (ii) sensing-centric design, and (iii) EH-centric design. In each strategy, we aim to jointly design the transmit communication precoders and sensing precoder (which also acts as both the energy beam for EH and the artificial noise for securing communication) subject to requirements corresponding to the transmit power budget, minimum achievable secrecy rate, minimum amount of harvested energy, and maximum beampattern similarity constraint. The optimization design problems are highly non-convex, and thus it is mathematically challenging to directly find the optimal solutions. To overcome this issue, we first derive convex inner approximations, and then utilize the sequential convex programming (SCP) approach to develop efficient iterative algorithms which converge at least to a locally optimal solution. Extensive numerical results are simulated to verify the optimality of developed algorithms and highlight the impact of sensing on the system secrecy rate and harvested energy.