In this paper, a novel laser spot tracking algorithm that incorporates the Kalman filter with the continuously adaptive Meanshift algorithm (Cam-Kalm) is proposed and employed in an underwater optical wireless communication (UOWC) system. Since the Kalman filter has the advantage of predicting the state information of the target spot based on its spatial motion features, the proposed algorithm can improve the accuracy and stability of the moving laser spot tracking. A 2 m optical wireless communication experimental system with auto-tracking based on a green laser diode (LD) is built to evaluate the tracking performance of different algorithms. Experimental results verify that the proposed algorithm outperforms conventional tracking algorithms in aspects of tracking accuracy, interference resistance, and response time. With the proposed Cam-Kalm algorithm, the experimental system can establish an effective communication link, while the maximum tracking speed is 20 mm/s given the forward-error-correction (FEC) threshold.
Read full abstract