Improving performance of distributed detection networks: an entropy-based optimization
Résumé
The reduction of communication costs is the key focus of distributed detection networks. On the contrary, their performance is lower compared to centralizednetworks, because the fusion center does not receive all the available information to make the final decision. The optimization of a parallel distributed network with N sensors leads to a set of 2N +N nonlinear equations, that can only be solved in particular cases, assuming statistical independence of the local observations, andfor small-sizednetworks. In order to reduce the number of equations to solve, Shannon’s entropy is used to select relevant sensors for the decision process. Once the relevant sensors are selected, an entropy-based criterion is used to optimize the network: local thresholds are determined, and a binary decision tree represents the fusion rule. This approach provides a fusion center using e6cient and simple procedures. These optimization techniques are applied to the distributed quantization problem: we consider a compromise between the broadcast bandwidth of the fusion center and quality of the decision system, in order to approximate centralized detection performance.