TY - GEN

T1 - A greedy algorithm for decentralized Bayesian detection with feedback

AU - Dong, Weiqiang

AU - Kam, Moshe

N1 - Funding Information:
This study was sponsored by the Office of Naval Research (ONR) under grant no. N00014-13-1-0733.
Publisher Copyright:
© 2016 IEEE.

PY - 2017/2/7

Y1 - 2017/2/7

N2 - We consider a decentralized binary detection architecture comprised of n local detectors (LDs) communicating their decisions to a Data Fusion Center (DFC). Each local detector (LD) declares preference for one of two hypotheses (H0 or H1) and transmits it to the DFC. The decision of the kth LD at time step t is utk. The DFC develops a global preference ut0 for one of hypotheses based on the vector of local decisions Ut while minimizing a Bayesian cost. The input to the kth LD at time step t is the observation ytk collected from the surveyed environment, and the previous global decision ut-10. Alhakeem and Varshney developed a person-by-person optimal (PBPO) solution to this problem, namely a PBPO procedure to calculate the global fusion rule and the local decision rules. However, their solution requires that at each time step 2n fusion rule equations and 2n local threshold equations be solved simultaneously. In this paper we suggest a suboptimal solution to the problem, based on independent local minimizations of similar Bayesian cost by each LD and by the DFC. To assess the cost of decentralization, the performance of this solution is compared to that of an architecture that processes all observations in one central location.

AB - We consider a decentralized binary detection architecture comprised of n local detectors (LDs) communicating their decisions to a Data Fusion Center (DFC). Each local detector (LD) declares preference for one of two hypotheses (H0 or H1) and transmits it to the DFC. The decision of the kth LD at time step t is utk. The DFC develops a global preference ut0 for one of hypotheses based on the vector of local decisions Ut while minimizing a Bayesian cost. The input to the kth LD at time step t is the observation ytk collected from the surveyed environment, and the previous global decision ut-10. Alhakeem and Varshney developed a person-by-person optimal (PBPO) solution to this problem, namely a PBPO procedure to calculate the global fusion rule and the local decision rules. However, their solution requires that at each time step 2n fusion rule equations and 2n local threshold equations be solved simultaneously. In this paper we suggest a suboptimal solution to the problem, based on independent local minimizations of similar Bayesian cost by each LD and by the DFC. To assess the cost of decentralization, the performance of this solution is compared to that of an architecture that processes all observations in one central location.

KW - Data fusion

KW - Decentralized detection detection

KW - Decision fusion

KW - Distributed detection

KW - Fusion with feedback

KW - Greedy algorithm

UR - http://www.scopus.com/inward/record.url?scp=85015241586&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015241586&partnerID=8YFLogxK

U2 - 10.1109/SARNOF.2016.7846756

DO - 10.1109/SARNOF.2016.7846756

M3 - Conference contribution

AN - SCOPUS:85015241586

T3 - 37th IEEE Sarnoff Symposium, Sarnoff 2016

SP - 202

EP - 207

BT - 37th IEEE Sarnoff Symposium, Sarnoff 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 37th IEEE Sarnoff Symposium, Sarnoff 2016

Y2 - 19 September 2016 through 21 September 2016

ER -