Physics-Based Deep Learning for Fiber-Optic Communication Systems

Autor: Christian Häger, Henry D. Pfister
Rok vydání: 2021
Předmět:
Signal Processing (eess.SP)
FOS: Computer and information sciences
Computer Science - Artificial Intelligence
Computer Networks and Communications
Computer science
Computer Science - Information Theory
Machine Learning (stat.ML)
02 engineering and technology
symbols.namesake
Statistics - Machine Learning
FOS: Electrical engineering
electronic engineering
information engineering
0202 electrical engineering
electronic engineering
information engineering
Electrical Engineering and Systems Science - Signal Processing
Electrical and Electronic Engineering
Nonlinear Schrödinger equation
Pointwise
Artificial neural network
business.industry
Information Theory (cs.IT)
Deep learning
Supervised learning
020206 networking & telecommunications
Filter (signal processing)
Backpropagation
Split-step method
Artificial Intelligence (cs.AI)
symbols
Artificial intelligence
Gradient descent
business
Algorithm
Zdroj: IEEE Journal on Selected Areas in Communications. 39:280-294
ISSN: 1558-0008
0733-8716
DOI: 10.1109/jsac.2020.3036950
Popis: We propose a new machine-learning approach for fiber-optic communication systems whose signal propagation is governed by the nonlinear Schr\"odinger equation (NLSE). Our main observation is that the popular split-step method (SSM) for numerically solving the NLSE has essentially the same functional form as a deep multi-layer neural network; in both cases, one alternates linear steps and pointwise nonlinearities. We exploit this connection by parameterizing the SSM and viewing the linear steps as general linear functions, similar to the weight matrices in a neural network. The resulting physics-based machine-learning model has several advantages over "black-box" function approximators. For example, it allows us to examine and interpret the learned solutions in order to understand why they perform well. As an application, low-complexity nonlinear equalization is considered, where the task is to efficiently invert the NLSE. This is commonly referred to as digital backpropagation (DBP). Rather than employing neural networks, the proposed algorithm, dubbed learned DBP (LDBP), uses the physics-based model with trainable filters in each step and its complexity is reduced by progressively pruning filter taps during gradient descent. Our main finding is that the filters can be pruned to remarkably short lengths-as few as 3 taps/step-without sacrificing performance. As a result, the complexity can be reduced by orders of magnitude in comparison to prior work. By inspecting the filter responses, an additional theoretical justification for the learned parameter configurations is provided. Our work illustrates that combining data-driven optimization with existing domain knowledge can generate new insights into old communications problems.
Comment: 15 pages, 11 figures, submitted to IEEE J. Sel. Areas Commun., code available at https://github.com/chaeger/LDBP, extension of arXiv:1710.06234(1), arXiv:1804.02799(1), arXiv:1901.07592(2)
Databáze: OpenAIRE
Externí odkaz: