Self-Sustainable DF-IRS-Aided Integrated Sensing, Communication, and Power Transfer
Abstract
This paper establishes a self-sustainable dual-functional intelligent reflecting surface (S-DF-IRS)-assisted integrated sensing, communication, and power transfer (ISCPT) system that synergistically combines target sensing, user communication, and wireless energy harvesting. Unlike existing studies, we conduct a systematic comparative investigation of three fundamental operational modes: power splitting (PS), time splitting (TS), and element splitting (ES). Our objective is to maximize the weighted sum of sensing mutual information and communication rate, subject to practical constraints. To this end, we develop a unified optimization framework and an efficient iterative algorithm based on alternating optimization, integrating successive convex approximation and semidefinite relaxation to jointly optimize active and passive beamforming, as well as mode-specific parameters. Extensive simulations demonstrate that the proposed hardware architecture achieves significant performance gains over a non-sensing S-IRS baseline, with improvements of 34.4% for PS, 32.7% for TS, and 31.2% for ES schemes, respectively. Among the three modes, ES consistently delivers the best performance, followed by PS and then TS. Furthermore, the optimized PS strategy yields up to 11.2% performance gain over fixed-ratio PS, highlighting the importance of dynamic resource allocation. It is also shown that ES outperforms both MRT and MRC strategies, and that a higher number of reflection elements further enhances system performance. These findings provide practical guidelines for the design of next-generation IRS-aided ISCPT systems.