Minimizing Angle Estimation Mean Squared Error Using an Improved Artificial Rabbits Optimization Algorithm

In RIS-assisted wireless communication systems, accurate angle information is essential for beamforming and channel estimation. However, in dynamic channel environments, variations in the Angle of Departure (AoD) can degrade the performance of conventional estimation methods. To address this issue, this paper proposes a RIS phase optimization method based on an Improved Artificial Rabbits Optimization (iARO) algorithm to enhance the angle tracking accuracy of the Extended Kalman Filter (EKF). First, a RIS-assisted dynamic angle tracking model is established, where the real and imaginary parts of the complex channel gain together with the AoD are considered as state variables, and EKF is applied for recursive estimation. Then, taking the mean square error (MSE) of the EKF angle estimation as the objective function, the iARO algorithm is employed to globally optimize the phase configuration of RIS elements, thereby improving the sensitivity of the received signal to angle variations. The proposed algorithm incorporates opposition-based learning initialization, global-best neighborhood search, gravitational guidance, and an iterative restart strategy, which enhance global search capability and convergence stability.