Introduction of neural networks in the decentralized detection problems

The problem of decentralized detection appears when various sensors are used simultaneously to collect observations from multiple points apart from each other. In this case, some difficulties such as transmission errors, very high transmission rates or memory requirement may occur when a centralized detection is used. To override these limitations detection structures involving two processing levels have been introduced: the decentralized detection. Unfortunately the solutions proposed to the decentralized problem require many calculations. Basing their approach on the similarity between detection problems and neural networks the authors propose a neural approach to solve the decentralized detection problem with fewer calculations