High-throughput hardware deployment of pruned neural network based nonlinear equalization for 100-Gbps short-reach optical interconnect

Hardware implementation of neural network based nonlinear equalizers will encounter tremendous challenges due to a high-throughput data stream and high computational complexity for 100-Gbps short-reach optical interconnects. In this Letter, we propose a parallel pruned neural network equalizer for high-throughput signal processing and minimized hardware resources. The structure of a time-interleaved neural network equalizer with a delay module is deployed in a field programmable gate array with advanced pruned algorithms, demonstrating significant bit error rate reduction for 100-Gbps real-time throughput with 200 parallel channels. Moreover, the dependence of processing throughput, hardware resources, and equalization performance is investigated, showing that over 50% resource reduction without performance degradation can be achieved with the pruning strategy. (C) 2021 Optical Society of America

Automated summary

The hardware implementation of neural network based nonlinear equalizers for 100-Gbps short-reach optical interconnects faces challenges due to high-throughput data stream and complexity. A parallel pruned neural network equalizer is proposed for high-throughput signal processing and minimized hardware resources. The study shows a significant reduction in bit error rate for 100-Gbps real-time throughput with 200 parallel channels, while achieving over 50% resource reduction without performance degradation through pruning strategy.