Collaboration in distributed hypothesis testing with quantized prior probabilities

Joong Bum Rhim, Lav R. Varshney, Vivek K. Goyal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The effect of quantization of prior probabilities in a collection of distributed Bayesian binary hypothesis testing problems over which the priors themselves vary is studied. In a setting with fusion of local binary decisions by majority rule, optimal local decision rules are discussed. Quantization is first considered under the constraint that agents employ identical quantizers. A method for design is presented that exploits an equivalence to a single-agent problem with a different likelihood function, the optimal quantizers are thus different than in the single-agent case. Removing the constraint of identical quantizers is demonstrated to improve performance. A method for design is presented that exploits an equivalence between agents having diverse K-level quantizers and agents having identical (3K-2)-level quantizers.

Original languageEnglish (US)
Title of host publicationProceedings - DCC 2011
Subtitle of host publication2011 Data Compression Conference
Pages303-312
Number of pages10
DOIs
StatePublished - May 12 2011
Externally publishedYes
Event2011 Data Compression Conference, DCC 2011 - Snowbird, UT, United States
Duration: Mar 29 2011Mar 31 2011

Publication series

NameData Compression Conference Proceedings
ISSN (Print)1068-0314

Other

Other2011 Data Compression Conference, DCC 2011
CountryUnited States
CitySnowbird, UT
Period3/29/113/31/11

ASJC Scopus subject areas

  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Collaboration in distributed hypothesis testing with quantized prior probabilities'. Together they form a unique fingerprint.

Cite this