Efficient Probabilistic Filtering of Data Subject to Channel Noise Under Local Pooling Conditions

The widespread use of wireless sensor networks has led to the solution of one of the important problems associated with assessing their state in conditions of damaged communication channels. Known evaluation algorithms are based on a centralized or distributed data filtering structure. This paper proposes a distributed algorithm of three-stage Kalman filtering, which combines and processes data locally. The proposed design of the algorithm made it possible to reduce and stabilize the mean square error of the filter and limit it uniformly from above.