This paper addresses the robust filter design problem for a class of Linear Repetitive Processes (LRPs) operating over amplify-and-forward (AF) relay networks subject to autocorrelated Rayleigh fading. To improve communication quality in long-distance transmissions and mitigate the inherent limitations of wireless sensor nodes, a three-terminal AF relay framework with autocorrelated Rayleigh fading under a two-dimensional (2-D) system setting is introduced. Unlike conventional approaches that decouple the fading parameters from the system state, this work defines a globally coupled correlation term that integrates both components, allowing direct incorporation into the filter design. A joint estimation framework is proposed, involving both the actual and an auxiliary filter. By employing mathematical induction, stochastic analysis, and a set of tunable parameters, tight upper bounds for the estimation error covariance are established and minimized through appropriate filter gain selection. Furthermore, the boundedness of the actual filter’s estimation error covariance is rigorously analyzed under mild parameter conditions. The effectiveness and applicability of the proposed filtering scheme are validated through its application to metal rolling process.

This paper addresses the problem of outlier-resistant estimation in a two-dimensional (2-D) Fornasini-Marchesini local state-space (FMLSS) systems with dead-zone-like censoring under asynchronous random delays. To mitigate trend outliers caused by extreme noise, a dead-zone-like censoring mechanism is introduced, where the truncation region is constrained by predetermined left and right truncation thresholds. Additionally, considering the delay phenomenon in networked systems, a new model is developed to describe asynchronous random delays under the assumption of a bounded maximum delay. To handle the incomplete information resulting from these delays, an innovative approach is proposed, using a Tobit fictitious observation sequence. By rearranging the fictitious observation sequence, the original delayed Tobit fictitious observation system is transformed into a multi-channel delay-free rearranged Tobit observation system. Based on this, an (N+1)-step extended KF is proposed, which employs adjustable scale parameters to derive a tight upper bound for the estimation error covariance. This bound is then minimized to design the estimation gain. Furthermore, the boundedness of the estimation error covariance is analyzed in the 2-norm sense. Finally, the effectiveness of the proposed outlier-resistant estimation approach is demonstrated through a numerical example

This paper addresses the state estimation problem for a class of networked two-dimensional systems subject to autocorrelated multiplicative noises and multi-step observation delays. The autocorrelated multiplicative noises are characterized as a linear system corrupted by additive white noises, and the multi-step observation delays are described by the instantaneous and delayed output equations. The globally coupled correlation terms are first introduced to deal with the autocorrelated multiplicative noises, and then a novel observation reconstruction approach is proposed to tackle the difficulty induced by multi-step observation delays. By utilizing the space equivalence verification, it is shown that the reconstructed delay-free observation sequence retains the same information as the original delayed observation sequence. The purpose of this paper is to design an iterative coupled estimator based on the globally coupled correlation terms and the reconstructed observation sequence. By means of the mathematical induction technique, it is demonstrated that the proposed estimator preserves the property of unbiasedness. Moreover, the estimator gains are obtained by minimizing the estimation error covariance, and some sufficient conditions are derived to guarantee the existence of an upper bound for the estimation error covariance in the mean-square sense. Finally, a numerical example is provided to verify the effectiveness of the proposed estimation scheme.

This paper addresses the distributed recursive filtering problem for a class of time-varying multi-rate nonlinear systems which are subject to fading channels, sensor resolutions and quantization effects. For the purpose of prolonging the transmission distance, the amplify-and-forward (AaF) relay is deployed in the wireless communication channel. The iterative method is employed to convert the underlying multi-rate system into a single-rate one. Before entering communication network for transmission, measurement outputs are quantized using a logarithmic quantizer. Moreover, a new equivalent delay-free measurement sequence is defined by the measurement reorganization technology to tackle random transmission delays. Our main aim is to design a distributed recursive filter such that an upper bound for the filtering error covariance is determined and subsequently minimized by properly choosing the filtering gains. Next, the performance of the proposed filtering algorithm is analyzed through establishing a sufficient condition for the boundedness of the filtering error covariance. Finally, a numerical simulation is used to testify the applicability and effectiveness of the developed filtering algorithm.

Abstract This study examines the H 2 filtering problem for sampled‐data systems that are subject to measurement delays and packet dropouts. The phenomenon of packet dropout is different for each observation component and is described as an independent and identically distributed Bernoulli process, which is more general than the scalar packet dropout case. The reorganized observation technique is introduced herein to address the time delays in the observations. An optimal filter is constructed using a deterministic continuous‐time system with finite jumps, where the filter gains are determined by integrating a specific continuous‐time Riccati equation with finite jumps, based on a suitable algebraic Lyapunov equation. Compared with the discrete‐time filter, the optimal filter can provide the state estimates of the measurement sampling intervals and measurement sampling moments. Two simulation examples are explored to demonstrate the effectiveness of the proposed method.

The linear quadratic regulation (LQR) problem for discrete-time networked control systems (NCSs) is investigated in this article. The difference from most previous works is that input delay and packet losses occur simultaneously in the communication channel connecting the controller to the actuator. Moreover, the data packet dropout is modeled as a time-homogeneous Markov process which poses challenges due to the temporal correlation. The contributions of this article are twofold. First, by applying the maximum principle involving Markov jumps and delay, the linear quadratic optimal control problem in finite horizon is solved and the solution is given in terms of a forward and a backward stochastic difference equations (FBSDEs-M). Second, under a basic assumption, the infinite horizon optimal control problem is solved and the necessary and sufficient condition for mean square stabilization is given in terms of the solutions to coupled algebraic Riccati-type equations with Markov jumps. The presented results are new to the best of our knowledge since there is no existing work that tackles delay and Markovian packet dropouts simultaneously. It is a generalization of the previous work in which the packet dropouts is modeled as an independent identically distributed Bernoulli process.

Modelling and handling epistemic uncertainty with belief function theory, different ways to learn classification forests from evidential training data are explored. In this paper, multiple base classifiers are learned on uncertain training subsets generated by training instance random sampling approach. For base classifier learning, with the tool of evidential likelihood function, gini impurity intervals of uncertain datasets are calculated for attribute splitting and consonant mass functions of labels are generated for leaf node prediction. The construction of gini impurity based belief binary classification tree is proposed and then compared with C4.5 belief classification tree. For base classifier combination strategy, both evidence combination method for consonant mass function outputs and majority voting method for precise label outputs are discussed. The performances of different proposed algorithms are compared and analysed with experiments on VCI Balance scale dataset.

This paper is interested in the receding horizon estimation problem for the networked control systems with packet losses. Different from the scalar data losses case, we introduce a diagonal matrix to represent the phenomenon of packet losses, where each element of the diagonal matrix is a binary stochastic variable and indicates the arrival of the corresponding measurement component. That means different observation components have different packet loss probability. A batch form and a recursive form for the receding horizon estimation are proposed by minimizing a new cost function that includes two terminal weighting terms. Using the derived condition, the stability of the proposed receding horizon estimation is proved. Finally, a numerical example is provided for illustration.

This paper mainly investigates the output feedback tracking control and stabilization for network control systems with packet losses. Both finite-horizon and infinite-horizon cases are considered. We introduce an augmented state consisting of the system state and the reference trajectory to address the problem in finite-horizon. And then the optimal closed-loop control, the minimum cost function and the optimal estimator are offered by solving a set of difference Riccati equations. In infinite-horizon, we provide a necessary and sufficient condition for the stabilization of the system, the optimal control and the optimal cost function are establishing based on a set of algebraic Riccati equations.

Geometric constraint problem can be transformed to an optimization problem. We can solve the problem with niche improved particle swarm. Classical particle swarm optimization is likely to be trapped into local minima as well as premature. A niche improved particle swarm optimization (NIPSO) based on niche theory was developed. After the update of the particle velocity and position, the outlier particle was identified in the NIPSO by comparing the niche number of every particle, with which the crossover and selection operators were employed sequent for those particles, whose personal best values were less than that of the outlier particle. The experiment shows that it can improve the geometric constraint solving efficiency and possess better convergence property than the compared algorithms.