Sinusoid Det JASA 1980

Sinusoid Det JASA 1980 - Detection of sinusoids in ocean...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Detection of sinusoids in ocean acoustic background noise W. S. Hodgkiss and V. C. Anderson University of California at San Diego, Marine Physical Laboratory, Scripps Institution of Oceanography, San Diego, California 92152 (Received 19 January 1979; accepted for publication 26 September 1979) A common receiver structure for the detection of sinusoidal signals obscured by a noise background is the filter, square, and integrate processor. For this processor, an interpretation of postdetection integration gain is given in terms of a low-pass faltering of the noise power time series. Although a processing gain of 1.5 dB per integration interval doubling is predicted, subsequent analysis on the highly colored envelope spectrum of ambient ocean acoustic noise data in the kilohertz region shows that its nonstationarity significantly alters realizable processing gains for integration intervals of tens to hundreds of seconds. PACS numbers: 43.60.Gk, 43.30.Sf, 92.10.Vz, 43.30.Ng INTRODUCTION A common approach to the problem of detecting sinus- oids buried in an ambient ocean noise background has been to perform short-term coherent spectral analysis via FFT techniques, then incoherently average these power spectra over several consecutive blocks of data. Total incoherent integration intervals on the order of tens to hundreds of seconds are frequently employed. Performance prediction analyses typically associate a 1.5 dB "processing gain" for every doubling of this post- detection integration interval. Implicit in such predic- tions is stationarity of the ambient noise background. Recent experiments with a high gain, medium fre- quency, acoustic array have demonstrated significant noise nonstationarity at the output of extremely narrow beams for periods greater than 15 s. ' Of interest here will be the implications of such nonstationarity on pro- cessor performance in terms of realizable processing gain for various integration intervals. The organization of this paper is as follows: Sec. I defines the filter, square, and integrate processor in the context of FFT-based digital signal processors. A detectability measure is introduced and is intuitively related to the comparison of a given frequency domain "bin" with the average level of adjacent bins. Then, Sec. II provides an interpretation of postdetection inte- gration gain in terms of a low-pass filtering or smooth- ing of the noise power time series. The envelope spec- trum is introduced as a characterization of the noise power fluctuation about its mean value. Sec. III follows with an example of ambient ocean noise and the impli- cations its nonstationarity has on realizable processing gain. Lastly, a summary is provided in Sec. IV. I. SINUSOlD
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/10/2012 for the course ECE 254 taught by Professor Hodgekiss during the Fall '11 term at UCSD.

Page1 / 6

Sinusoid Det JASA 1980 - Detection of sinusoids in ocean...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online