Mitigation red-shift effect in multi-color LEDs-based visible light communication with signal-dependent noise

Visible light communication (VLC) has been proposed as a potential solution for future wireless communication networks to address the limited availability of the radio-frequency spectrum. A significant challenge with VLC is that the thermal parameters of light-emitting diodes (LEDs), particularly the junction temperature, are highly dependent on their operating conditions. This is because as the junction temperature increases, the power spectral densities of LEDs shift and broaden into the red region at different rates in the visible spectrum, known as the red-shift effect. Furthermore, signal-dependent noise, an inherent characteristic of the physical layer in VLC, and the red-shift effects can change the probability density function of the received signal. Additionally, fluctuations in signal intensity caused by signal-dependent noise result in errors during the signal detection process. For these reasons, the conventional maximum-likelihood (ML) receiver does not perform optimally. In this context, this study is the first to examine the red-shift effect in color-shift keying modulation-based VLC (VLC-CSK) systems in the presence of signal-dependent noise. This study proposes an optimal ML receiver that addresses both the red-shift effect and signal-dependent noise while maintaining satisfactory bit error rate (BER) performance. The proposed technique employs the received signal power of each color channel to correct the constellation diagram distorted by the red-shift effect prior to signal detection. Here, the Monte Carlo simulation results demonstrate that the red-shift effect and signal-dependent noise significantly impact the BER performance and lead to substantial degradation. Our proposed method enhances BER performance at high signal-to-noise ratios, offering considerable potential for designing efficient indoor VLC-CSK networks, and exhibits notable effectiveness in terms of computational complexity.