Performance Analysis of Generalized Metropolis-Hastings Algorithm over Mobile Wireless Sensor Networks

In the last years, we could have witnessed significant progress in wireless sensor networks in terms of many aspects such as hardware components, communication and routing protocols, etc. Lately, the attention of both the academy and the industry has been more and more focused on the mobile variant of this technology. Even though mobility brings many advantages, a lot of challenges related to sensor networks are still not solved. In this paper, we address the generalized Metropolis- Hastings algorithm for distributed averaging, whose goal is to overcome negative factors affecting the sensor readings as well as suppressing data redundancy. We test the algorithm with a varied mixing parameter over stationary edge-Markovian evolving graphs in order to identify the optimal configuration of the analyzed algorithm.