研究目的
To demonstrate a phosphorescent white LED CAP 64QAM VLC system utilizing deep neural network (DNN) and LMS linear equalization (LE) for high-speed data transmission with improved performance in mitigating nonlinear distortion.
研究成果
The experimental demonstration shows that a phosphorescent white LED CAP 64QAM VLC system utilizing DNN and LE can achieve a data rate of 2.4Gb/s with BER below the 7% FEC limit, indicating DNN+LE as a promising solution for indoor high-speed VLC systems.
研究不足
The study is limited to indoor free space transmission over a 1.1-m distance. The performance of DNN+LE is compared only with LE and volterra NLE+LE, and the system's effectiveness in other scenarios or with other equalization methods is not explored.
1：Experimental Design and Method Selection:
The experiment involves a CAP 64QAM VLC system using DNN and LE for post equalization to mitigate nonlinear and linear distortions respectively.
2：Sample Selection and Data Sources:
The system uses a phosphorescent white LED for transmission and a commercial PIN photodiode for reception.
3：List of Experimental Equipment and Materials:
Includes an arbitrary waveform generator, pre-equalization circuit, electrical amplifier (EA), phosphorescent white LED, lens, blue filter, PIN photodiode, and digital storage oscilloscope.
4：Experimental Procedures and Operational Workflow:
The process involves signal generation, pre-equalization, transmission through LED, reception via PIN photodiode, amplification, and offline signal processing including DNN and LE equalization.
5：Data Analysis Methods:
BER performance is measured versus different bias currents and signal Vpp, and spectra are compared before and after equalization.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
