A distributed estimation method for sensor networks based on Pareto optimization
2012 (English)In: Decision and Control (CDC), 2012 IEEE 51st Annual Conference on, IEEE , 2012, 775-781 p.Conference paper (Refereed)
A novel distributed estimation method for sensor networks is proposed. The goal is to track a time-varying signal that is jointly measured by a network of sensor nodes despite the presence of noise: each node computes its local estimate as a weighted sum of its own and its neighbors' measurements and estimates and updates its weights to minimize both the variance and the mean of the estimation error by means of a suitable Pareto optimization problem. The estimator does not rely on a central coordination: both parameter optimization and estimation are distributed across the nodes. The performance of the distributed estimator is investigated in terms of estimation bias and estimation error. Moreover, an upper bound of the bias is provided. The effectiveness of the proposed estimator is illustrated via computer simulations and the performances are compared with other distributed schemes previously proposed in the literature. The results show that the estimation quality is comparable to that of one of the best existing distributed estimation algorithms, guaranteeing lower computational cost and time.
Place, publisher, year, edition, pages
IEEE , 2012. 775-781 p.
, IEEE Conference on Decision and Control. Proceedings, ISSN 0191-2216
Electrical Engineering, Electronic Engineering, Information Engineering
IdentifiersURN: urn:nbn:se:kth:diva-118791DOI: 10.1109/CDC.2012.6426731ScopusID: 2-s2.0-84874275025OAI: oai:DiVA.org:kth-118791DiVA: diva2:608392
51st IEEE Conference on Decision and Control, CDC 2012; Maui, HI;10 December 2012 through 13 December 2012
QC 201303042013-02-272013-02-272013-03-04Bibliographically approved