Additive decomposition of one-dimensional signals using Transformers

TL;DR

This paper introduces a novel Transformer-based method for decomposing one-dimensional signals into multiple components, demonstrating high accuracy on synthetic data and offering a promising deep learning approach to traditional signal analysis tasks.

Contribution

The work presents the first application of Transformers to additive signal decomposition, enabling effective separation of signal components with a deep learning model trained on synthetic data.

Findings

High accuracy in component decomposition on synthetic signals

Effective separation of constant, smooth, textured, and noise components

Potential for improved pre-processing in data analysis pipelines

Abstract

One-dimensional signal decomposition is a well-established and widely used technique across various scientific fields. It serves as a highly valuable pre-processing step for data analysis. While traditional decomposition techniques often rely on mathematical models, recent research suggests that applying the latest deep learning models to this problem presents an exciting, unexplored area with promising potential. This work presents a novel method for the additive decomposition of one-dimensional signals. We leverage the Transformer architecture to decompose signals into their constituent components: piece-wise constant, smooth (low-frequency oscillatory), textured (high-frequency oscillatory), and a noise component. Our model, trained on synthetic data, achieves excellent accuracy in modeling and decomposing input signals from the same distribution, as demonstrated by the experimental results.