Abstract: The security performance of the communication system degrades dramatically when the covariance- based Channel State Information (CSI) is imperfect at the transmitter. To overcome this problem, a robust Artificial Noise (AN) aided transmit method is proposed. The objective is to jointly design the transmit beamforming vector and the AN covariance with the imperfect covariance-based CSI at the transmitter, such that the Worst-Case Secrecy Rate (WCSR) of the system is maximized. The secrecy rate maximization problem is non-convex. Due to the intractability, this problem is recast into a series of SemiDefinite Programs (SDPs) using the SemiDefinite Relaxation (SDR) technique and the Lagrange duality. Simulation results demonstrate that the proposed method provides substantial performance improvements over the existing method.