3 November 2018 in Circuits, Systems, and Signal Processing
9 January 2020 in IEEE Access
Censored measurements frequently occur in network systems involving censored sensors or saturated sensors. In addition, unreliable network characteristics can produce random measurement delays during signal transmission. In this paper, we investigate the state estimation problem for network systems with the simultaneous appearance of the aforementioned two measurement uncertainties. The occurrences of two random measurement phenomena are described by two Bernoulli random variables in which the censored variable is dependent on the delay variable. The probability of the process signal being uncensored is calculated by the local approximations using a priori and a posteriori of the state estimation. Then, a novel measurement model that incorporates both the censoring random matrix and the signal delay is established. Based on this model, an optimal recursive estimation method is proposed for systems with specified two uncertainties by making use of an innovation analysis approach and a weighted conditional expectation formula. The superior performance of our proposed method is verified through a typical oscillator simulation example.
5 December 2022 in IEEE Transactions on Circuits and Systems II: Express Briefs
Censored measurements are frequently occurring phenomena in sensor systems with the characteristics of output dead-zone or sensor saturation. In censoring sensor network systems, random transmission loss is inevitable, which results in the two measurement uncertainties occurring simultaneously. In this brief, we address this problem and attempt to present an efficient solution. First, the random transmission losses are described by a set of Bernoulli random variables, and then the censored random matrix and translation transformation are used to modify the traditional Tobit censored measurement model, which includes calculating the probability that the latent measurement is not censored when the transmission is lost and is only pure noise. Then, an optimal recursive filtering algorithm for a system with the two uncertain measurements is derived by using the innovation analysis method and the full expectation rule. The effectiveness and superiority of the algorithm is verified by simulating an oscillator model and comparing with other existing algorithms.
1 August 2020 in Electronics Letters
The development of particle filters for emerging data transmission applications is essential to address network‐related issues. In this Letter, a robust particle filter algorithm for non‐linear systems with one‐step randomly delayed measurements and outlier corrupted measurement noise is investigated. To address the issue of measurement outliers, a Student's ‐t distribution is utilised to model the measurement noise. For the random delay uncertainties, a Bernoulli random variable is employed, based on which the likelihood function is transformed into an exponential multiplication form. The filter state estimation is obtained by Bayes's theorem and using variational Bayesian approach, the latent random variables and Bernoulli random variable are estimated together. The superior performance of the proposed method is verified through a typical example by comparison with existing particle filters that consider random measurement delay or heavy‐tailed measurement noise.