Title :
Nonparametric decentralized sequential detection
Author_Institution :
Dept. of Electr. Eng., Hawaii Univ., Honolulu, HI, USA
fDate :
29 Jun-4 Jul 1997
Abstract :
We consider a decentralized sequential detection problem with a set of sensors and a fusion center. Each sensor receives information from inputs and possibly other sensors at discrete times and transmits summary information to a fusion center which processes the summary information by performing a sequential test to make a decision on one of two hypotheses. The work discussed differs from previous work by Veervalli, Basar and Poor (1993) in that the conditional densities given each hypothesis are unknown. Information about making good decisions is learned from observing real data and employing reinforcement learning procedures
Keywords :
feedforward neural nets; learning (artificial intelligence); sensor fusion; sequences; signal detection; conditional densities; feedforward neural network; fusion center; nonparametric decentralized sequential detection; real data observations; reinforcement learning procedures; sensors; sequential test; summary information transmission; Bayesian methods; Cost function; Dynamic programming; Learning; Neural networks; Performance evaluation; Sensor fusion; Sensor phenomena and characterization; Sensor systems; Testing;
Conference_Titel :
Information Theory. 1997. Proceedings., 1997 IEEE International Symposium on
Conference_Location :
Ulm
Print_ISBN :
0-7803-3956-8
DOI :
10.1109/ISIT.1997.613465