Robust design of sparse array-based sensing-aided physical layer secure transmission in mixed far-field and near-field environments

Abstract

When the eavesdropper and the legitimate receiver are located at the same angle but different distances in mixed far-field and near-field environments, existing results on sensing-aided physical layer security (PLS) based on far-field only assumption would fail. In this paper, a sparse array-based robust sensing-aided PLS transmission algorithm is investigated in mixed far-field and near-field environments to address the above issue. The proposed scheme adopts a transceiverinterleaved sparse array architecture, which expands the physical aperture and improves spatial resource utilization while reducing cost. To consider uncertainties in the sensing stage, the Cramér–Rao Bound (CRB) of the target parameters is derived to construct a channel-domain error uncertainty set. Subsequently, a robust optimization problem that maximizes the worst-case secrecy rate (WCSR) is formulated. The non-convex problem is solved by employing bisection search and the S-Procedure lemma. Simulation results demonstrate that the proposed algorithm actively widens the beam nulls and effectively interferes with all potential eavesdroppers to deal with location uncertainties, thereby reliably guaranteeing the PLS performance even in harsh sensing environments.