Simultaneous Perturbation Stochastic Approximation-Based Consensus Algorithm for Decentralized Cooperative Online Estimation

Network systems, that consist of multiple nodes capable of sensing, processing and communicating, estimate some unknown parameters of a dynamic process more accurately due to communication between network nodes. Supervisory Control And Data Acquisition (SCADA) system is a good example of system with different measured parameters. In this paper a decentralized cooperative online estimation with Simultaneous Perturbation Stochastic Approximation-based (SPSA-based) consensus is considered. The proposed algorithm is a combination of two well-studied parts: Simultaneous Perturbation Stochastic Approximation and Local Voting Protocol. The numeric example of SPSA-based consensus algorithm applied to SCADA system is presented.