Implementation of Multiple Transmitter Tracking Mechanism Based on DeepSORT for 2D MIMO Optical Camera Communications

This paper presents an enhanced 2D MIMO Optical Camera Communication (OCC) system that integrates a multi-transmitter tracking mechanism enhanced by the DeepSORT algorithm to address real-time communication challenges in dynamic environments. The system employs YOLOv11 for accurate LED transmitter detection and a deep learning-based decoder to improve decoding efficiency, resulting in a lower Bit Error Rate compared to conventional methods. The DeepSORT algorithm enables the simultaneous tracking of up to 10 LED transmitters with high accuracy, reaching 99.6% multiple object tracking accuracy and 99.3% multiple object tracking precision. Experimental results demonstrate a throughput of up to 15.360 kbps for a 16x16 LED matrix, which showcases the system’s scalability and suitability for IoT and smart city applications. In addition, the deep learning-based decoder improves the signal-to-noise ratio under challenging lighting conditions, which further enhances system robustness. These advancements highlight the system’s potential for stable, high-performance communication in dynamic and noisy environments and offer a promising solution for future optical wireless communication technologies.