Distributed decision-making with bayesian and neural network approaches
Dimitrios K. Bougoulias, Stelios C. A. Thomopoulos · 1991
The problem of Optimal Multi Sensor Decision Fusion, is first considered. The fusion center receives data from various distributed sensors and combines them into a final decision. Assuming that the sensor distributions are independent from each other conditioned on each hypothesis, a general proof is provided that the optimal decision scheme that maximizes the probability of detection at the fusion, for a fixed false alarm probability, consists of a Neyman Pearson (N-P) test or a randomized N-P test at the fusion and likelihood ratio tests at the sensors. Next, DIGNET, a self-organizing Artificial Neural Network (ANN) is presented that exhibits deterministically reliable behavior to noise interference, when the noise does not exceed a pre-specified level of tolerance. The complexity of the proposed ANN, in terms of neuron requirements versus stored patterns, increases linearly with the number of stored patterns and their dimensionality. The self-organization of the DIGNET is based on the idea of competative generation and elimination of attraction wells in the pattern space. This ANN can be used for both pattern recognition and classification. Application of DIGNET in character recognition and unknown signal detection is examined. Finally, the DIGNET is used for Multi Sensor Decision Fusion. One network is used at each sensor performing detection of unknown signals. Another network used at the fusion center performs classification and decides which is the correct hypothesis.