Dual-functional Radar Communication enabled Physical Layer Security under CRB based Statistical Imperfect CSI Model

The integrated sensing and communication (ISAC) gives the new paradigm for the secure communication design. In this paper, the dual-functional radar communication (DFRC) enabled physical layer security (PLS) is investigated. Specifically, we propose the novel Cramér-Rao bound (CRB) based statistical imperfect channel state information (CSIC) model, which integrates the radar CRB into the statistical imperfect CSI model based on the second-order moment. Further, the CSIC based PLS optimization model is proposed, where the expectation of the secrecy rate is maximized under the PLS requirement and the radar CRB based constraint. The corresponding CSIC semidefinite relaxation (CSIC-SDR) based iteration algorithm is developed to achieve the beamforming design with the aid of one-dimension search of the optimal CRB based sensing measure. Numerical results demonstrate that the proposed CSIC-SDR outperforms the norm-bound CSI based and circularly symmetric complex Gaussian distribution CSI based beamforming designs in terms of the PLS performance.