TY  - JOUR
T1  - Optimizing OFDM Modulation in OCC Systems: A Comparative Analysis of AI-Based Optical Channel Equalization Models
AU  - Kim, Jeong Eun 
AU  - Jang, Yeong Min 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2025
DA  - 2025/1/1
DO  - 10.7840/kics.2025.50.7.1153
KW  - OFDM
KW  - Optical Camera Communication (OCC)
KW  - Signal Denoising
KW  - AI-based Equalizer
AB  - Optical camera communication (OCC) is an emerging wireless technology that utilizes image sensors to receive data from modulated light sources. However, conventional OCC systems suffer from low data rates and high bit error rates (BER) due to environmental noise, motion blur, and ambient light interference. To address these challenges, this study integrates orthogonal frequency division multiplexing (OFDM) with AI-driven optical channel equalization, employing a BiLSTM-based equalizer to mitigate signal degradation and enhance transmission reliability. By treating pixel rows as transmission units, the system achieves a data rate of 6.2 kbps at a 3-m distance with a BER of 2.88×10−2, demonstrating substantial improvements over conventional OCC methods. A comparative analysis of deep learning-based equalizers shows that BiLSTM outperforms ResNet and BiGRU in denoising performance, as evidenced by superior signal reconstruction metrics, including the lowest MSE (0.72%) and RMSE (7.24%), as well as the highest R2 (86.75%) and PCC (0.95). The system is implemented and optimized using Python-based scripts, enabling real-time processing and embedded deployment. These findings highlight the potential of AI-enhanced OFDM equalization for next-generation OCC systems, providing a robust and high-speed optical wireless communication framework adaptable to real-world applications.