Secure Beamforming for NOMA-ISAC With System Imperfections

Transmitting communication and sensing signals simultaneously poses a potential security risk. Existing security non-orthogonal multiple access and integration of sensing and communication (NOMA-ISAC) schemes may fail to achieve secure communications when some legitimate users situate at close angles with respect to BS. To address this issue, we investigate a secure beamforming and power allocation based on user clustering to ensure NOMA-ISAC security. Different from conventional secure NOMA-ISAC schemes, we apply a beamforming vector to the users at the same cluster to suppress the interferences from other clusters, facilitating the decode of NOMA signals. We formulate a secure beamforming and power allocation optimization problem with imperfect channel state information (CSI) and successive interference cancellation (SIC). To solve the non-convex optimization problem, we decompose it into two sub-problems with the alternating optimization (AO) method. Then semidefinite relaxation (SDR) is employed to transform the sub-problems into convex optimization problem, and iterative rank minimization (IRM) is employed to approximately achieve the non-convex rank-one constraints. Numerical results demonstrate the effectiveness of the proposed method.