Parametric Bootstrapping of Array Data with A Generative Adversarial Network

Since the number of independent array data snapshots is limited by the availability of real-world data, we propose a parametric bootstrap for resampling. The proposed parametric bootstrap is based on a generative adversarial network (GAN) following the generative approach to machine learning. For th...

Full description

Saved in:
Bibliographic Details
Main Authors: Gerstoft, Peter, Groll, Herbert, Mecklenbrauker, Christoph F.
Format: Conference Proceeding
Language:English
Subjects:
Acoustic arrays
Array signal processing
DOA estimation
Games
GAN
generative adversarial network
Generative adversarial networks
Generators
Hellinger distance
Machine learning
Oceans
sample covariance matrix
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Since the number of independent array data snapshots is limited by the availability of real-world data, we propose a parametric bootstrap for resampling. The proposed parametric bootstrap is based on a generative adversarial network (GAN) following the generative approach to machine learning. For the GAN model we chose the Wasserstein GAN with penalized norm of gradient of the critic with respect to its input (wGAN gp). The approach is demonstrated with synthetic and real-world ocean acoustic array data.
ISSN:2151-870X
DOI:10.1109/SAM48682.2020.9104371