Open AccessLinear estimators for networked systems with one‐step random delay and multiple packet dropouts based on prediction compensation
Author(s) -
Ma Jing,
Sun Shuli
Publication year - 2017
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
eISSN - 1751-9683
pISSN - 1751-9675
DOI - 10.1049/iet-spr.2016.0260
Subject(s) - estimator , network packet , computer science , moment (physics) , compensation (psychology) , filter (signal processing) , computational complexity theory , algorithm , mathematics , statistics , psychology , computer network , physics , classical mechanics , psychoanalysis , computer vision
This study is concerned with the linear estimation problem for networked systems with one‐step random delay and multiple packet dropouts. At each moment, the estimator may receive one or two data packets or nothing. The predictor of sensor measurement at the current instant is used as a compensator if the current measurement does not arrive at the estimator. Based on the developed compensation model, the optimal linear estimators including filter, predictor and smoother are proposed by the innovation analysis approach. Compared with the estimators based on the compensation of using the latest measurement previously received in the existing literatures, the proposed estimators have higher estimation accuracy and smaller computational burden. Simulation results show the effectiveness of the proposed algorithms.