A memoryless grouped-data nonparametric sequential detection procedure

M. M. Al-Ibrahim, Pramod Kumar Varshney

Research output: Chapter in Book/Entry/PoemConference contribution

Abstract

Summary form only given, as follows. A nonparametric sequential testing procedure for the detection of constant signal in additive symmetric noise is proposed and analyzed. The nonparametric test is obtained by first quantizing all the observations into their signs and then applying the Lee-Thomas modifed sequential procedure to the quantized observations or their sufficient statistic. The procedure is extended to a distributed system consisting of two local detectors and a global decision maker. A simple truncation scheme is considered and is shown to maintain the nonparametric property while efficiently limiting the random test duration. Numerical results are presented to illustrate the performance gain and to demonstrate the simplicity of the test structure.

Original languageEnglish (US)
Title of host publication1990 IEEE Int Symp Inf Theor
PublisherIEEE Computer Society
Pages7
Number of pages1
StatePublished - 1990
Event1990 IEEE International Symposium on Information Theory - San Diego, CA, USA
Duration: Jan 14 1990Jan 19 1990

Other

Other1990 IEEE International Symposium on Information Theory
CitySan Diego, CA, USA
Period1/14/901/19/90

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'A memoryless grouped-data nonparametric sequential detection procedure'. Together they form a unique fingerprint.

Cite this