Rethinking Covert ISAC With RIS and Imperfect CSI: How Invisible Can Communication Be Under Generalized Fading?
Abstract
This paper investigates covert integrated sensing and communication (ISAC) systems assisted by reconfigurable intelligent surfaces (RIS) under independent and identically distributed (i.i.d) generalized fading and imperfect channel state information (CSI). The main goal is to develop a robust joint beamforming and RIS design that maximizes radar sensing performance while satisfying communication quality-of-service and covert communication requirements in practical wireless environments. To this end, we derive a tractable approximation of the detection error probability at the warden by modeling the received signal power as a Gamma random variable. Thanks to this modeling, and under the considered statistical assumptions, the covertness constraint reduces to a simple transmit-power ceiling independent of the RIS configuration. Based on this result, we formulate a joint optimization problem to maximize the radar sensing signal-to-noise ratio while satisfying communication quality-of-service and covertness constraints. The resulting non-convex problem is solved using an alternating optimization framework combining semidefinite relaxation and fractional programming, and its convergence is theoretically established. Simulation results, conducted for the $\alpha $ - $\mu $ fading model as a representative example of generalized fading channels, validate the accuracy of the proposed analytical approximation and demonstrate that the proposed optimization framework effectively balances sensing performance, communication reliability, and covertness. In particular, while RIS improves sensing performance, it also increases signal leakage toward the warden, thereby tightening the covertness constraint, and imperfect CSI leads to diminishing gains as the number of RIS elements increases.