Secrecy rate maximization for hardware impaired untrusted relaying network with deep learning

This paper investigates the physical layer security (PLS) design of an untrusted relaying network where source node coexists with a multi-antenna eavesdropper (Eve). While communication relies on untrustworthy relay nodes to increase reliability, we aim protect confidentiality information against combined eavesdropping attacks performed by both and Eve. Considering hardware impairments (HIs), total power budget constraint for whole individual at each node, this presents novel approach jointly optimize beamformer transmit powers aiming maximizing average secrecy rate (ASR). To safeguard first cooperative phase, destination-aided jamming (DACJ) is employed, while second adjusted. The resultant optimization problem non-convex, suboptimal solution obtained through sequential parametric convex approximation (SPCA) method. In order prevent any failure due infeasibility, propose iterative initialization algorithm find feasible initial point original instead arbitrary point, as in conventional SPCA. satisfy low-latency one main key performance indicators (KPI) required beyond 5G (B5G) communications, computationally efficient data-driven developed exploiting deep learning model evaluate proposed scheme computational burden significantly reduced. Simulation results assess effect different system parameters ASR well effectiveness large-scale cases.