Quick and Easy Time Series Generation with Established Image-based GANs

Abstract

In the recent years Generative Adversarial Networks (GANs) have demonstrated significant progress in generating authentic looking data. In this work we introduce our simple method to exploit the advancements in well established image-based GANs to synthesise single channel time series data. We implement Wasserstein GANs (WGANs) with gradient penalty due to their stability in training to synthesise three different types of data; sinusoidal data, photoplethysmograph (PPG) data and electrocardiograph (ECG) data. The length of the returned time series data is limited only by the image resolution, we use an image size of 64x64 pixels which yields 4096 data points. We present both visual and quantitative evidence that our novel method can successfully generate time series data using image-based GANs.

References in corpus (2)

• Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
• EEG-GAN: Generative adversarial networks for electroencephalograhic (EEG) brain signals

Cited by in corpus (10)

• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• Conditional GAN for timeseries generation
• Generative adversarial networks in time series: A survey and taxonomy
• Generative Adversarial Networks: A Survey Towards Private and Secure Applications
• Synthetic-Neuroscore: Using A Neuro-AI Interface for Evaluating Generative Adversarial Networks
• GMM-Based Generative Adversarial Encoder Learning
• A Neuro-AI Interface for Evaluating Generative Adversarial Networks
• A Spectral Enabled GAN for Time Series Data Generation
• Synthesis of standard 12-lead electrocardiograms using two dimensional generative adversarial network
• EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision