Abstract:Photonics-assisted millimeter wave (MMW) can overcome the bandwidth bottleneck of electronic devices, but it also faces nonlinear damage caused by optoelectronic devices. Nonlinear equalization algorithms based on neural networks (NN) have gradually been introduced in the receiving end of 6G MMW communication, but the high computational complexity limits the real-time performance of the equalizer, and there is a tradeoff between the good equalization performance and low complexity. This paper proposes an Automatic Search Algorithm (ASA) based on deep reinforcement learning (DRL), which learns the functional relationship between equalizer structure parameters and optimization objectives through DRL, in order to quickly search for the optimal structure of various equalizers. The experiment demonstrated the W-band 16QAM discrete multi-tone (DMT) signal transmission in an intensity modulation and direct detection (IM/DD) system over a 50 m wireless link. We implement real-valued time-domain neural network equalizer (TDNNE), and complex-valued frequency-domain neural network equalizer (FDNNE), to solve the nonlinear issues respectively. Subsequently, ASA and greedy search algorithm were applied to TDNNE and FDNNE respectively, and the superiority of ASA was verified based on the bit error rate (BER) and the number of Multiply-Accumulate Operations (MACs) consumed by the equalizer. Finally, the experiment verified that using ASA to optimize the pruning of FDNNE can greatly reduce the complexity with almost no BER loss, while retaining the good equalization performance of NN equalizers and overcoming the disadvantage of high computational complexity. Thanks to our proposed complex-valued FDNNE based on DRL, the effective data rate reaches 11.4Gb/s with 0.5 dB receiver sensitivity improvement compared to VNLE and TDNNE. At the same time, after pruning optimization, the average number of MACs consumed by FDNNE at various input optical powers into photodiode (PD) reduced by 37%. We believe ASA is an important application of using DRL to optimize neural networks in the field of nonlinear equalization. It can efficiently explore the performance limits of equalizers and has great potential for high-precision reception in future 6G wireless communications.

Amplified radio-over-fiber system linearization using recurrent neural networks

Demonstration of photonic neural network for fiber nonlinearity compensation in long-haul transmission systems

Deep Learning Equalizer Connected with Viterbi-Viterbi Algorithm for PAM D-Band Radio over Fiber Link

Digital Predistortion Based Experimental Evaluation of Optimized Recurrent Neural Network for 5G Analog Radio Over Fiber Links

A Data-driven Digital Demodulator based on Deep Learning for Radio over Fiber Transmission System

Experimental Demonstration of Millimeter-Wave Radio-over-Fiber System with Convolutional Neural Network (CNN) and Binary Convolutional Neural Network (BCNN)

Low Computationally Complex Recurrent Neural Network for High Speed Optical Fiber Transmission

Implementation of HetNet architectures based on FSO, VLC, RoF, and photonics-based RF generation toward 6G applications

Efficient Deep Learning of Nonlinear Fiber-Optic Communications Using a Convolutional Recurrent Neural Network

81-GHz W-band 60-Gbps 64-QAM wireless transmission based on a dual-GRU equalizer

High-Speed Terahertz Band Radio-Over-Fiber System Using Hybrid Time-Frequency Domain Equalization

Fiber neural networks for the intelligent optical fiber communications

Nonlinearity Mitigation in a Fiber-Wireless Integrated System Based on Low-Complexity Autoencoder and BiLSTM-ANN Equalizer.

Frequency-Domain Complex-Valued NN Equalizer Based on Deep Reinforcement Learning for Photonics-Assisted W-Band 16QAM-DMT Signals System over 50 M

Deep Learning Interference Cancellation in Wireless Networks

Transformer-based Nonlinear Equalization for DP-16QAM Coherent Optical Communication Systems

A Machine Learning Aided Reference-Tone-Based Phase Noise Correction Framework for Fiber-Wireless Systems

Performance and complexity study of a neural network post-equalizer in a 638-nm laser transmission system through over 100-m plastic optical fiber

170-260 GHz Sub-THz Optical Heterodyne Analog Radio-over-Fiber Link for 6G Wireless System

Deep-learning-based Multi-User Framework for End-to-end Fiber-Mmw Communications.

W-band photonic-aided reconfigurable multichannel vector mm-wave transmission enabled by the digital subcarrier multiplexing technique