Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Robust Channel Estimation for Optical Wireless Communications Using Neural Network

Dianxin Luan, John Thompson

TL;DR

The paper tackles robust channel estimation for frequency-selective optical wireless channels by introducing a multi-branch adaptive neural estimator that selects among pre-trained InterpolateNet networks based on estimated delay spread. The framework uses a primary HDS-trained model to produce an initial estimate, then classifies the PDP to choose the most suitable LDS, MDS, or HDS network, achieving low complexity of $O(N_f N_s) + O(N_f \log N_f)$. Simulation results show NMSE and BER improvements over traditional LS/MMSE and non-adaptive neural baselines in indoor OWC scenarios, demonstrating robustness to SNR variations and channel dynamics. This work highlights the potential of PDP-guided neural estimators to enhance reliability of indoor OWC systems while maintaining computational efficiency, with offline training on a large channel dataset and publicly available code for replication.

Abstract

Optical Wireless Communication (OWC) has gained significant attention due to its high-speed data transmission and throughput. Optical wireless channels are often assumed to be flat, but we evaluate frequency selective channels to consider high data rate optical wireless or very dispersive environments. To address this for optical scenarios, this paper presents a robust channel estimation framework with low-complexity to mitigate frequency-selective effects, then to improve system reliability and performance. This channel estimation framework contains a neural network that can estimate general optical wireless channels without prior channel information about the environment. Based on this estimate and the corresponding delay spread, one of several candidate offline-trained neural networks will be activated to predict this channel. Simulation results demonstrate that the proposed method has improved and robust normalized mean square error (NMSE) and bit error rate (BER) performance compared to conventional estimation methods while maintaining computational efficiency. These findings highlight the potential of neural network solutions in enhancing the performance of OWC systems under indoor channel conditions.

Robust Channel Estimation for Optical Wireless Communications Using Neural Network

TL;DR

The paper tackles robust channel estimation for frequency-selective optical wireless channels by introducing a multi-branch adaptive neural estimator that selects among pre-trained InterpolateNet networks based on estimated delay spread. The framework uses a primary HDS-trained model to produce an initial estimate, then classifies the PDP to choose the most suitable LDS, MDS, or HDS network, achieving low complexity of . Simulation results show NMSE and BER improvements over traditional LS/MMSE and non-adaptive neural baselines in indoor OWC scenarios, demonstrating robustness to SNR variations and channel dynamics. This work highlights the potential of PDP-guided neural estimators to enhance reliability of indoor OWC systems while maintaining computational efficiency, with offline training on a large channel dataset and publicly available code for replication.

Abstract

Optical Wireless Communication (OWC) has gained significant attention due to its high-speed data transmission and throughput. Optical wireless channels are often assumed to be flat, but we evaluate frequency selective channels to consider high data rate optical wireless or very dispersive environments. To address this for optical scenarios, this paper presents a robust channel estimation framework with low-complexity to mitigate frequency-selective effects, then to improve system reliability and performance. This channel estimation framework contains a neural network that can estimate general optical wireless channels without prior channel information about the environment. Based on this estimate and the corresponding delay spread, one of several candidate offline-trained neural networks will be activated to predict this channel. Simulation results demonstrate that the proposed method has improved and robust normalized mean square error (NMSE) and bit error rate (BER) performance compared to conventional estimation methods while maintaining computational efficiency. These findings highlight the potential of neural network solutions in enhancing the performance of OWC systems under indoor channel conditions.

Paper Structure

This paper contains 7 sections, 10 equations, 3 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System settings and channel model
  3. Robust channel estimation for wireless optical channels
  4. Simulation Results
  5. NMSE performance on randomly generated channels
  6. BER performance on randomly generated channels
  7. Conclusion

Figures (3)

  • Figure 1: Example of a two-path optical wireless channel where the magnitude of the reflection path is tens dB less than that of the LOS path, but channel gain varies among these subcarriers.
  • Figure 2: NMSE performance of LS, MMSE, InterpolateNet trained with HDS conditions and the proposed method.
  • Figure 3: BER results of LS, MMSE, direct signal detection, HDS-Net and the proposed method over SNR from 15dB to 30dB.