Distributed and Recursive Parameter Estimation in Parametrized Linear State-Space Models - Computer Science > Distributed, Parallel, and Cluster ComputingReportar como inadecuado




Distributed and Recursive Parameter Estimation in Parametrized Linear State-Space Models - Computer Science > Distributed, Parallel, and Cluster Computing - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: We consider a network of sensors deployed to sense a spatio-temporal fieldand estimate a parameter of interest. We are interested in the case where thetemporal process sensed by each sensor can be modeled as a state-space processthat is perturbed by random noise and parametrized by an unknown parameter. Toestimate the unknown parameter from the measurements that the sensorssequentially collect, we propose a distributed and recursive estimationalgorithm, which we refer to as the incremental recursive prediction erroralgorithm. This algorithm has the distributed property of incremental gradientalgorithms and the on-line property of recursive prediction error algorithms.We study the convergence behavior of the algorithm and provide sufficientconditions for its convergence. Our convergence result is rather general andcontains as special cases the known convergence results for the incrementalversions of the least-mean square algorithm. Finally, we use the algorithmdeveloped in this paper to identify the source of a gas-leak diffusing sourcein a closed warehouse and also report numerical simulations to verifyconvergence.



Autor: S. Sundhar Ram, V. V. Veeravalli, A. Nedic

Fuente: https://arxiv.org/







Documentos relacionados