This paper proposes a movable antenna (MA)-aided hybrid beamforming scheme with a sub-connected structure for multi-user communications. The scheme allows multiple movable sub-arrays to independently adjust their positions within local regions. The goal is to maximize the system sum rate by jointly optimizing the digital beamformer, analog beamformer, and positions of sub-arrays, subject to constraints on unit modulus, finite movable regions, and power budget. Due to the non-concave/non-convex objective function and highly coupled variables, the problem is challenging to solve. To address this, the authors employ fractional programming and develop an alternating optimization framework combining Lagrange multipliers, penalty method, and gradient descent. Numerical results show that the proposed MA-aided hybrid beamforming scheme significantly improves the sum rate compared to its fixed-position antenna (FPA) counterpart. Moreover, with sufficiently large movable regions, the sub-connected MA array outperforms the fully-connected FPA array. The system model considers an MA-aided downlink multi-user multi-input single-output (MU-MISO) system with a multi-antenna base station (BS) and K single-antenna users. The BS uses a sub-connected hybrid beamforming structure consisting of movable uniform planar arrays (UPAs). Each UPA is connected to a set of phase shifters (PSs) and a dedicated RF chain, allowing collective movement of the antennas. The signal model includes transmit and receive field-response vectors, path-response matrix, and channel vectors. The problem formulation aims to maximize the sum rate by jointly optimizing the positions of UPAs, analog beamformer, and digital beamformer. The proposed solution involves a low-complexity algorithm based on fractional programming and alternating optimization. The algorithm decomposes the problem into three subproblems: digital beamformer design, analog beamformer design, and MA position design. The digital beamformer is optimized using a convex quadratic optimization approach, while the analog beamformer is optimized using the penalty method. The MA positions are optimized using a gradient descent method. Numerical results demonstrate the effectiveness of the proposed scheme, showing that it outperforms both FPA-based systems and the upper bound on sum rate under certain conditions. The results highlight the benefits of movable antennas in enhancing communication performance through spatial channel exploitation.This paper proposes a movable antenna (MA)-aided hybrid beamforming scheme with a sub-connected structure for multi-user communications. The scheme allows multiple movable sub-arrays to independently adjust their positions within local regions. The goal is to maximize the system sum rate by jointly optimizing the digital beamformer, analog beamformer, and positions of sub-arrays, subject to constraints on unit modulus, finite movable regions, and power budget. Due to the non-concave/non-convex objective function and highly coupled variables, the problem is challenging to solve. To address this, the authors employ fractional programming and develop an alternating optimization framework combining Lagrange multipliers, penalty method, and gradient descent. Numerical results show that the proposed MA-aided hybrid beamforming scheme significantly improves the sum rate compared to its fixed-position antenna (FPA) counterpart. Moreover, with sufficiently large movable regions, the sub-connected MA array outperforms the fully-connected FPA array. The system model considers an MA-aided downlink multi-user multi-input single-output (MU-MISO) system with a multi-antenna base station (BS) and K single-antenna users. The BS uses a sub-connected hybrid beamforming structure consisting of movable uniform planar arrays (UPAs). Each UPA is connected to a set of phase shifters (PSs) and a dedicated RF chain, allowing collective movement of the antennas. The signal model includes transmit and receive field-response vectors, path-response matrix, and channel vectors. The problem formulation aims to maximize the sum rate by jointly optimizing the positions of UPAs, analog beamformer, and digital beamformer. The proposed solution involves a low-complexity algorithm based on fractional programming and alternating optimization. The algorithm decomposes the problem into three subproblems: digital beamformer design, analog beamformer design, and MA position design. The digital beamformer is optimized using a convex quadratic optimization approach, while the analog beamformer is optimized using the penalty method. The MA positions are optimized using a gradient descent method. Numerical results demonstrate the effectiveness of the proposed scheme, showing that it outperforms both FPA-based systems and the upper bound on sum rate under certain conditions. The results highlight the benefits of movable antennas in enhancing communication performance through spatial channel exploitation.