Using logistic regression classification for mitigating high noise-ratio advisement light-panel in rolling-shutter...

Area of Science:

• Optoelectronics
• Machine Learning
• Communication Systems

Background:

• Visible Light Communication (VLC) systems offer a promising alternative for high-speed wireless data transmission.
• Traditional demodulation schemes in VLC can be susceptible to noise and interference, limiting performance.
• Machine learning (ML) presents an opportunity to enhance signal processing and error correction in communication systems.

Purpose of the Study:

• To propose and experimentally demonstrate a novel VLC system incorporating an ML algorithm.
• To evaluate the performance of the ML algorithm against traditional demodulation techniques.
• To assess the effectiveness of the ML approach in mitigating noise in VLC systems.

Main Methods:

• Development of a VLC system utilizing a light-panel as the transmitter and an image sensor as the receiver.
• Implementation of a machine learning algorithm for data demodulation and error correction.
• Comparative analysis of the ML algorithm's bit error rate (BER) against a traditional demodulation scheme under varying noise conditions.

Main Results:

• The proposed ML algorithm demonstrated superior performance compared to the traditional demodulation scheme.
• Significant improvements in bit error rate (BER) were observed even at high noise-ratio (NR) levels.
• The system successfully transmitted data using a light-panel display under challenging noisy conditions.

Conclusions:

• The integration of ML algorithms offers a robust solution for enhancing the reliability of VLC systems.
• The proposed ML-based VLC system shows significant potential for practical applications requiring high data integrity in noisy environments.
• Machine learning provides a powerful tool for overcoming performance limitations in visible light communication technology.