Energy Efficiency Optimization for Integrated Sensing and Communications-aided Full Duplex MIMO System with Imperfect CSI

Next-generation wireless networks are expected to deliver exceptional sensing functionalities with communication systems, enabling advanced applications such as intelligent automation, real-time data processing, and enhanced virtual environments. This paper examines an unconventional integrated sensing and communication (ISAC) with full-duplex (FD) multiple-input multiple-output (MIMO) system where a dualfunction radar and communication (DFRC) base station (BS) performs multiple target sensing and shares the same resources with multiple uplink (UL) and downlink (DL) communication users. To establish a green FD-ISAC architecture, we primarily focus on the robust energy efficiency (EE) maximization problem, which involves joint optimization of receive and transmit beamforming for multiple users and radar beamforming for multiple targets under norm-bounded channel state information (CSI) error and other quality of service (QoS) requirements. We employ an iterative algorithm based on successive convex approximation (SCA) and semi-definite programming (SDP) to achieve near-optimal resource allocation. The proposed approach is evaluated against benchmark methods, including exhaustive search, perfect CSI, and half-duplex schemes. Numerical results demonstrate the superiority of the proposed FD-ISAC system over conventional ISAC frameworks, even in the presence of residual self-interference (RSI) and imperfect CSI.