Journal of intelligence and knowledge engineering.
Artificial Noise-Aided Secrecy Beamforming Optimization Method for MISO Wiretap Channels
Ruixin Li
Abstract
Physical layer security technologies, which do not rely on complex key management and utilize the inherent physical characteristics of wireless channels to achieve secure information transmission, are a critical means of ensuring the confidentiality of future wireless communications. Artificial noise-aided beamforming design in Multiple-Input Single-Output (MISO) wiretap channels offers significant advantages in combating passive eavesdropping; however, existing optimization methods still suffer from shortcomings in relaxation accuracy, convergence efficiency, and robustness. This paper addresses the secrecy rate maximization problem in artificial noise-aided MISO wiretap channels. It constructs a non-convex optimization model incorporating total power constraints, null-space orthogonality constraints, sparse directional constraints, and dual robustness constraints. Simulation results demonstrate that the proposed solution framework integrating alternating optimization and semidefinite relaxation can significantly improve the system's secrecy rate and achieve faster convergence. Even in scenarios with channel estimation errors and eavesdropper location uncertainty, the algorithm maintains stable secrecy performance, verifying its superior robustness.