Title :
A non-fluctuating target model for McDaniel´s reverberation
Author :
Abraham, Douglas A. ; Gu, Ming
Author_Institution :
Appl. Res. Lab., Pennsylvania State Univ., University Park, PA, USA
Abstract :
Active sonar reverberation has often been observed to have probability tails heavier than the traditionally assumed Rayleigh distribution. Many other physics-based and phenomenological statistical models have been proposed to represent such observed reverberation, a few of which have natural target models. One that has shown promise in having a minimal parameterization yet a broad enough scope to adequately represent a wide variety of observed non-Rayleigh reverberation is McDaniel´s (1990) model. In this paper a statistical model is developed for a non-fluctuating target in McDaniel´s reverberation exploiting the non-central gamma distribution. The probability density function and cumulative distribution function are derived and used to determine likelihood ratio and locally optimal detector non-linearities and also theoretical receiver operating characteristic curves
Keywords :
gamma distribution; matched filters; maximum likelihood detection; reverberation; sonar detection; sonar signal processing; McDaniel´s reverberation; active sonar reverberation; cumulative distribution function; likelihood ratio; locally optimal detector nonlinearities; minimal parameterization; noncentral gamma distribution; nonfluctuating target; nonfluctuating target model; probability density function; probability tails; receiver operating characteristic curves; statistical model; Detectors; Distribution functions; Laboratories; Matched filters; Probability density function; Random variables; Reverberation; Rician channels; Sonar detection; Statistical distributions;
Conference_Titel :
OCEANS 2000 MTS/IEEE Conference and Exhibition
Conference_Location :
Providence, RI
Print_ISBN :
0-7803-6551-8
DOI :
10.1109/OCEANS.2000.881739