This paper addresses a secure multiple-input multiple-output (MIMO) communication system enhanced by movable antennas (MAs). The system consists of a base station (BS) equipped with MAs, a legitimate information receiver (IR), and an eavesdropper (Eve). The primary objective is to maximize the secrecy rate (SR) by jointly optimizing the transmit precoding (TPC) matrix, artificial noise (AN) covariance matrix, and MAs' positions, subject to maximum transmit power and minimum distance constraints. To solve this non-convex problem, a block coordinate descent (BCD) method combined with the majorization-minimization (MM) algorithm is proposed. The SR is reformulated using the minimum mean square error (MMSE) method, and the optimal TPC matrix and AN covariance matrix are derived using the Lagrangian multiplier method. The MM algorithm is then applied to iteratively optimize each MA's position while keeping others fixed. Simulation results demonstrate the effectiveness of the proposed algorithms and show that the MA-aided system significantly outperforms conventional fixed-position antenna (FPA)-based systems in terms of security performance. The increase in the transmit region size, number of antennas, number of paths, and transmit power enhances the security of the MA-aided system, and accurate field-response information (FRI) is crucial for its performance.This paper addresses a secure multiple-input multiple-output (MIMO) communication system enhanced by movable antennas (MAs). The system consists of a base station (BS) equipped with MAs, a legitimate information receiver (IR), and an eavesdropper (Eve). The primary objective is to maximize the secrecy rate (SR) by jointly optimizing the transmit precoding (TPC) matrix, artificial noise (AN) covariance matrix, and MAs' positions, subject to maximum transmit power and minimum distance constraints. To solve this non-convex problem, a block coordinate descent (BCD) method combined with the majorization-minimization (MM) algorithm is proposed. The SR is reformulated using the minimum mean square error (MMSE) method, and the optimal TPC matrix and AN covariance matrix are derived using the Lagrangian multiplier method. The MM algorithm is then applied to iteratively optimize each MA's position while keeping others fixed. Simulation results demonstrate the effectiveness of the proposed algorithms and show that the MA-aided system significantly outperforms conventional fixed-position antenna (FPA)-based systems in terms of security performance. The increase in the transmit region size, number of antennas, number of paths, and transmit power enhances the security of the MA-aided system, and accurate field-response information (FRI) is crucial for its performance.