Share Email Print

Proceedings Paper

Performance comparison between conventional direction-of-arrival algorithm and stereausis network based on cochlea model
Author(s): Yaoliang Zhang; Li Yin; Wei Liu
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In this paper, we present experimental results comparing the conventional direction-of-arrival (DOA) algorithm and the stereausis network algorithm for the purpose of performing acoustic DOA estimation. The comparison would be focused on robustness, complexity and aperture size. Robustness is important because the acoustic signatures can vary significantly under different environmental conditions. Low complexity is also important because the DOA algorithm will be used in real time and systems need to identify a target instantly. Aperture size is another important issue in target positioning because fewer sensors could lead to small size and be located more freely. We show the DOA results of different algorithms on a simulated signal with given SNRs (Signal to Noise Ratio) and discuss issues such as robustness with respect to noise, computational complexity, and aperture sizes. The stereausis network algorithm could perform well under low SNR environment. When the SNR is above -5db, the accuracy is almost not affected. The complexity of computation of stereausis algorithm is much smaller than temporal delay algorithm because of the absence of neural delays, which means it has no time delay units. Therefore, stereausis algorithm could resolve the direction with less time cost compared to conventional time-delay methods so that it exhibits high real time feature. In addition, conventional algorithms usually require complex sensors array, which means the aperture size can't be small, while the stereausis algorithm could be implemented through only two sensors, which means a smaller aperture size.

Paper Details

Date Published: 30 October 2009
PDF: 6 pages
Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74951P (30 October 2009); doi: 10.1117/12.832863
Show Author Affiliations
Yaoliang Zhang, Institute of Acoustics (China)
Li Yin, Institute of Acoustics (China)
Wei Liu, Institute of Acoustics (China)

Published in SPIE Proceedings Vol. 7495:
MIPPR 2009: Automatic Target Recognition and Image Analysis
Tianxu Zhang; Bruce Hirsch; Zhiguo Cao; Hanqing Lu, Editor(s)

© SPIE. Terms of Use
Back to Top