Search

This paper studies the distributed dimensionality reduction fusion estimation problem for cyber-physical systems with limited bandwidth in presence of eavesdroppers. Since wireless communication is implemented by broadcasting, the eavesdroppers can collude to collect the data through anther communication networks. To protect data privacy, based on the physical processes and local estimation error covariance (EEC) matrix, an insertion method of artificial noise (AN) is developed such that only eavesdroppers’ fusion EEC becomes worse. Meanwhile, the fusion center needs to decode the received signal due to the noise interference, while the successful decoding probability varies with signal to noise ratio. Subsequently, some criteria for the selection probabilities and the successful decoding probabilities are given to guarantee the effectiveness of the AN insertion strategy. Moreover, a sufficient condition of the designed AN power is derived to guarantee the confidentiality. Simulation examples are given to show the effectiveness of the proposed methods.

Vehicle state, including location and motion information, plays an essential role on the Internet of Vehicles (IoV). Accurately obtaining the system state information is the premise of realizing precise control. However, the statistics of system process noise are often unknown due to the complex physical process. It is challenging to estimate the system state when the process noise statistics are unknown. This paper proposes a new adaptive high-degree unscented Kalman filter based on the improved Sage–Husa algorithm. First, the traditional Sage–Husa algorithm is improved using a high-degree unscented transform. A noise estimator suitable for the high-degree unscented Kalman filter is obtained to estimate the statistics of the unknown process noise. Then, an adaptive high-degree unscented Kalman filter is designed to improve the accuracy and stability of the state estimation system. Finally, the target tracking simulation results verify the proposed algorithm’s effectiveness.

In this work, we study finite-horizon multiple-sensor scheduling for general scalar Gauss-Markov systems, extending previous results where only a class of systems are considered. The scheduling objective is to minimize the terminal estimation error covariance. Only one sensor can transmit its measurement per time instant and each sensor has limited energy. Through building a comparison function and solving its monotone intervals, an efficient algorithm is designed to construct the optimal schedule.

A filter based on characteristic functions is developed in this paper, to fit to a class of non-Gaussian dynamical systems, which state models and measurement models are all nonlinear and multi-dimensional. The new filter overcomes limitations and expands the application of this kind of filter, which is proved to just fit to one special kind of systems with multi-dimensional linear state models and one-dimensional nonlinear measurement models. Firstly, the filter using characteristic function is introduced and its limitation is analysed. Then, we design the new filter to fit to nonlinear states and multi-dimensional measurements. Thirdly, the matrix format of performance index is presented to match to the new filter gain, and the weighting function vector is given to ensure the uniform boundedness of such a performance index. Finally, the new filter gain can be obtained by minimizing this performance index, and the process of filtering design is accomplished. Simulation examples are given to illustrate the effectiveness of the proposed filter design scheme.

This paper designs a novel filter based on characteristic function for multidimensional observation systems, essentially extending the proposed filter, which just fitted to one-dimensional observations. For the dynamic model from grinding process cylindrical workpieces, this filter could result in enhanced and incremental productivity and quality control in manufacturing processes. In the processing of the filter design, a new form of filter will be given to adapt to the multidimensional observations, the matrix format of performance index will be designed to fit to matrix format of filter gain, the selecting range of the weighting function vector will be given to ensure the uniform boundedness of the designed performance index, and the filter gain can be obtained by minimizing the performance index. Finally, we illustrate the effectiveness of the proposed method by simulation examples in the field of the target tracking and the grinding process cylindrical workpieces.

Data fusion for nonlinear systems is one of the challenging topics in state estimation and target tracking recently. We study decentralized cubature Kalman fusion in this paper. Cubature Kalman filter (CKF) is a more effective method than the conventional nonlinear filters, such as extended Kalman filter (EKF) and unscented Kalman filter (UKF). For most of the practical cases, there are correlative between process and measurement noises (Correlation I) and among measurement noises (Correlation II). So, it is more attractive to design fusion algorithms based on the CKF for the systems with complex correlated noises. Firstly, a cubature Kalman filter with correlation I (CKF-CN) is derived. Secondly, by introducing the EKF with correlated noises (EKF-CN) and its information filter EIF-CN, the CKF-CN is embedded in the EIF-CN framework to get a cubature information filter with correlated noises (CIF-CN). Consequently, a square-root cubature Kalman filter with noise correlation I (SCKF-CN) and the associated information filter SCIF-CN are presented to improve computational performance. Finally, based on the proposed SCIF-CN and matrix diagonalization, a decentralized nonlinear fusion algorithm is proposed for the multisensor system with Correlation I and Correlation II. Simulation examples are demonstrated to validate the proposed filters and fusion algorithms.

Support vector regression algorithm is widely used in fault diagnosis of rolling bearing. A new model parameter selection method for support vector regression based on adaptive fusion of the mixed kernel function is proposed in this paper. We choose the mixed kernel function as the kernel function of support vector regression. The mixed kernel function of the fusion coefficients, kernel function parameters, and regression parameters are combined together as the parameters of the state vector. Thus, the model selection problem is transformed into a nonlinear system state estimation problem. We use a 5th-degree cubature Kalman filter to estimate the parameters. In this way, we realize the adaptive selection of mixed kernel function weighted coefficients and the kernel parameters, the regression parameters. Compared with a single kernel function, unscented Kalman filter (UKF) support vector regression algorithms, and genetic algorithms, the decision regression function obtained by the proposed method has better generalization ability and higher prediction accuracy.