Recursive Filtering Under Probabilistic Encoding-Decoding Schemes: Handling Randomly Occurring Measurement Outliers

This article focuses on the recursive filtering problem for networked time-varying systems with randomly occurring measurement outliers (ROMOs), where the so-called ROMOs denote a set of large-amplitude perturbations on measurements. A new model is presented to describe the dynamical behaviors of RO...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on cybernetics Vol. 54; no. 6; pp. 3378 - 3391
Main Authors: Zou, Lei, Wang, Zidong, Dong, Hongli, Yi, Xiaojian, Han, Qing-Long
Format: Journal Article
Language:English
Published: United States IEEE 01.06.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Boundedness analysis
Coding
Covariance
Electronic mail
Error analysis
Filter design (mathematics)
Measurement uncertainty
Outliers (statistics)
parameter-dependent recursive (PDR) filtering
Performance degradation
Pollution measurement
probabilistic encoding–decoding (PED) scheme
Probabilistic logic
Probability theory
randomly occurring measurement outliers (ROMOs)
Recursive functions
Scalars
Stochastic processes
Time-varying systems
Upper bound
Upper bounds
ISSN:2168-2267, 2168-2275, 2168-2275
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article focuses on the recursive filtering problem for networked time-varying systems with randomly occurring measurement outliers (ROMOs), where the so-called ROMOs denote a set of large-amplitude perturbations on measurements. A new model is presented to describe the dynamical behaviors of ROMOs by using a set of independent and identically distributed stochastic scalars. A probabilistic encoding-decoding scheme is exploited to convert the measurement signal into the digital format. For the purpose of preserving the filtering process from the performance degradation induced by measurement outliers, a novel recursive filtering algorithm is developed by using the active detection-based method where the "problematic" measurements (i.e., the measurements contaminated by outliers) are removed from the filtering process. A recursive calculation approach is proposed to derive the time-varying filter parameter via minimizing such the upper bound on the filtering error covariance. The uniform boundedness of the resultant time-varying upper bound is analyzed for the filtering error covariance by using the stochastic analysis technique. Two numerical examples are presented to verify the effectiveness and correctness of our developed filter design approach.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2168-2267
2168-2275
2168-2275
DOI:10.1109/TCYB.2023.3234452