Diffusion LMS-based distributed detection over adaptive networks

We consider the problem of distributed detection, where a set of nodes are required to decide between two hypotheses based on their measurements. In diffusion implementations, nodes communicate with their neighbors and no fusion center is needed. In previous work we proposed a distributed detection scheme which was based on diffusion least-squares techniques. In this work we consider the case where nodes utilize diffusion LMS techniques instead. The proposed detector is capable of tracking changes in the active hypothesis. We analyze the performance of the detector, and provide simulation results comparing with other cooperation schemes.