近日，上海交通大学的义理林及其研究团队取得一项新进展。经过不懈努力，他们提出一种作为光纤通信线性补偿新基准的可学习数字信号处理方法。相关研究成果已于2024年8月13日在国际知名学术期刊《光：科学与应用》上发表。

本文提出了一种基于可学习视角的高效数字信号处理（DSP）设计思路，称为可学习DSP（LDSP）。LDSP复用传统DSP模块，将整个DSP视为深度学习框架，并基于反向传播算法从全局角度自适应地优化DSP参数。该方法不仅树立了线性DSP性能的新标杆，也为非线性DSP设计提供了重要的基准。

与传统的超参数优化DSP相比，实验证明，在1600km光纤传输后，结合LDSP和基于扰动的非线性补偿算法，400Gb/s信号的Q因子实现了约1.21dB的显著提升。得益于可学习模型，LDSP能够以低复杂度自适应地学习最佳配置，降低了对初始参数的依赖。

所提方法实现了符号速率的DSP，以较小的误码率（BER）为代价，相比传统的每符号2个样本处理，复杂度降低了48%。研究人员相信，LDSP代表了一种新的、高效的DSP设计范式，有望在光通信的各个领域引起广泛关注。

据悉，下一代光纤传输技术的兴趣激增，推动了数字信号处理（DSP）方案的发展，这些方案兼具高性能和低复杂性，成本效益高。然而，作为非线性补偿方法的基准，传统DSP采用逐块模块化设计进行线性补偿，在补偿后可能会表现出残余的线性效应，从而限制了非线性补偿性能。

附：英文原文

Title: Learnable digital signal processing: a new benchmark of linearity compensation for optical fiber communications

Author: Niu, Zekun, Yang, Hang, Li, Lyu, Shi, Minghui, Xu, Guozhi, Hu, Weisheng, Yi, Lilin

Issue&Volume: 2024-08-13

Abstract: The surge in interest regarding the next generation of optical fiber transmission has stimulated the development of digital signal processing (DSP) schemes that are highly cost-effective with both high performance and low complexity. As benchmarks for nonlinear compensation methods, however, traditional DSP designed with block-by-block modules for linear compensations, could exhibit residual linear effects after compensation, limiting the nonlinear compensation performance. Here we propose a high-efficient design thought for DSP based on the learnable perspectivity, called learnable DSP (LDSP). LDSP reuses the traditional DSP modules, regarding the whole DSP as a deep learning framework and optimizing the DSP parameters adaptively based on backpropagation algorithm from a global scale. This method not only establishes new standards in linear DSP performance but also serves as a critical benchmark for nonlinear DSP designs. In comparison to traditional DSP with hyperparameter optimization, a notable enhancement of approximately 1.21dB in the Q factor for 400Gb/s signal after 1600km fiber transmission is experimentally demonstrated by combining LDSP and perturbation-based nonlinear compensation algorithm. Benefiting from the learnable model, LDSP can learn the best configuration adaptively with low complexity, reducing dependence on initial parameters. The proposed approach implements a symbol-rate DSP with a small bit error rate (BER) cost in exchange for a 48% complexity reduction compared to the conventional 2 samples/symbol processing. We believe that LDSP represents a new and highly efficient paradigm for DSP design, which is poised to attract considerable attention across various domains of optical communications.

DOI: 10.1038/s41377-024-01556-5

Source: https://www.nature.com/articles/s41377-024-01556-5