Extreme-Learning-Machine-Based Noniterative and Iterative Nonlinearity Mitigation for LED Communication Systems

Authors

Dawei Gao, University of WollongongFollow
Qinghua Guo, University of WollongongFollow
Jun Tong, University of WollongongFollow
Nan Wu
Jiangtao Xi, University of WollongongFollow
Yanguang Yu, University of WollongongFollow

RIS ID

143459

Publication Details

D. Gao, Q. Guo, J. Tong, N. Wu, J. Xi & Y. Yu, "Extreme-Learning-Machine-Based Noniterative and Iterative Nonlinearity Mitigation for LED Communication Systems," IEEE Systems Journal, 2020.

Abstract

This article concerns receiver design for light-emitting diode (LED) communications, where the LED nonlinearity can severely degrade the performance of the system. We propose extreme learning machine (ELM)-based noniterative and iterative receivers to effectively handle the LED nonlinearity and memory effects. For the iterative receiver design, we also develop a data-aided receiver, where data are used as virtual training sequence in ELM training. It is shown that the ELM-based receivers significantly outperform conventional polynomial-based receivers. Iterative receivers can achieve huge performance gain compared to noniterative receivers, and the data-aided receiver can reduce training overhead considerably. This article can also be extended to radio frequency communications, e.g., to deal with the nonlinearity of power amplifiers.