The Decision Fusion in the Wireless Network with Fading Channels

In wireless sensor networks, the problem of channel fading is attracting more and more attention. In this paper, we consider to get optimal local sensor compression rules that minimize the Bayes cost under the fading channels if the fusion rule and the probabilities of transmission error of fading channels are given. It doesn't require conditional independence between sensor data, so it is quite general and fit for the most distributed sensor network decision fusions under fading channels. The designed numerical examples in this paper illustrate the decision performance by ROC curves.