Sorry, you need to enable JavaScript to visit this website.

A deep network for single-snapshot direction of arrival estimation

Citation Author(s):
Peter Gerstoft, Emma Ozanich, Haiqiang Niu
Submitted by:
Emma Ozanich
Last updated:
28 October 2019 - 10:56am
Document Type:
Poster
Document Year:
2019
Event:
Presenters Name:
Peter Gerstoft
Paper Code:
182

Abstract 

Abstract: 

This paper examines a deep feedforward network for beamforming with the single--snapshot Sample Covariance Matrix (SCM). The Conventional beamforming formulation, typically quadratic in the complex weight space, is reformulated as real and linear in the weight covariance and SCM. The reformulated SCMs are used as input to a deep feed--forward neural network (FNN) for two source localization. Simulations demonstrate the effect of source incoherence and performance in a noisy tracking example. The FNN beamformer is experimentally tested on the Swellex96 experiment S95 source tow with a loud interferer.

up
0 users have voted:

Dataset Files

conference_poster_6.pdf

(376)