On Average Consensus Algorithm over Mobile Wireless Sensor Networks Modelled as Stationary Markovian Evolving Graphs

Mobile wireless sensor networks find application in various areas due to their specific character. However, their operation is affected by many negatives factors and therefore modern applications are equipped with complementary data aggregation mechanisms to supress negatives effects. In this paper, our attention is focused on five frequently applied weight models of the average consensus algorithm for distributed averaging over mobile wireless sensor networks modelled as stationary Markovian evolving graphs with a different size and a varying probabiltiy of edge formation. We use the mean square error over the iterations as a metric to evaluate the performance of the analyzed weight models and identify the weight model with the highest performance.