In:
The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 87, No. S1 ( 1990-05-01), p. S155-S155
Abstract:
The detection of transients is a special case of nonstationary time series. It encompasses the detection of transient sonar signals, including active sonar returns. The detection of active sonar returns is generally treated as a matched-filter problem. The matched filter requires total a priori knowledge of the signal which, ideally, is stored in the receiver in the form of a “replica.” Since both the sonar target and the propagation channel distort the signal, and the time of arrival is usually not known, this replica of the signal is never known exactly, and therefore the matched filter is always applied suboptimally in practice. In this paper, the application of polyspectra to the general problem of the detection of a deterministic transient signal in stationary but not necessarily Gaussian noise is examined. It is applicable to both transient and active sonar detection since the exact configuration of the signal does not need to be known a priori. This, coupled with the fact that there is no need to assume Gaussianity on the part of the stochastic part gives the method great power. Detection statistics are computed for the method and compared to those for the matched filter. Also, results using synthetic data will be shown.
Type of Medium:
Online Resource
ISSN:
0001-4966
,
1520-8524
Language:
English
Publisher:
Acoustical Society of America (ASA)
Publication Date:
1990
detail.hit.zdb_id:
1461063-2
Bookmarklink