Title :
A distributed M-ary hypothesis testing problem with correlated observations
Author :
Tang, Zhuang-Bo ; Pattipati, Krishna R. ; Kleinman, David L.
Author_Institution :
Dept. of Electr. & Syst. Eng., Connecticut Univ., Storrs, CT, USA
Abstract :
A distributed M-ary hypothesis testing problem with correlated observations is considered. In this problem, each of a number of N subordinate decision makers (DMs) transmits one of a prespecified set of M0 messages based on his or her own data to a primary DM, who, in turn, combines the messages with his or her own data to make the final decision. The necessary conditions for the optimal decision rules of the DMs are derived, and some properties of the decision rules are investigated. A nonlinear Gauss-Seidel iterative algorithm is developed to solve for the local (optimal) decision rules, and its monotonic convergence to the local optimum is established. The computational issues associated with the algorithm are discussed. The algorithm is illustrated by several examples, and implications for distributed organization design are pointed out
Keywords :
convergence; decision theory; iterative methods; probability; correlated observations; decision theory; distributed M-ary hypothesis testing; distributed organization design; local optimum; monotonic convergence; nonlinear Gauss-Seidel iterative algorithm; optimal decision rules; subordinate decision makers; Algorithm design and analysis; Command and control systems; Convergence; Costs; Gaussian processes; Iterative algorithms; System testing; Systems engineering and theory;
Conference_Titel :
Decision and Control, 1989., Proceedings of the 28th IEEE Conference on
Conference_Location :
Tampa, FL
DOI :
10.1109/CDC.1989.70177
