20 December 2021 in IEEE Transactions on Signal Processing
Parameter uncertainties and measurement outliers unavoidably exist in a real linear system. Such uncertainties and outliers make the true joint state-measurement distributions (induced by the true system model) deviate from the nominal ones (induced by the nominal system model) so that the performance of the optimal state estimator designed for the nominal model becomes unsatisfactory or even unacceptable in practice. The challenges are to quantitatively describe the uncertainties in the model and the outliers in the measurements, and then robustify the estimator in a right way. This article studies a distributionally robust state estimation framework for linear systems subject to parameter uncertainties and measurement outliers. It utilizes a family of distributions near the nominal one to implicitly describe the uncertainties and outliers, and the robust state estimation in the worst case is made over the least-favorable distribution. The advantages of the presented framework include: 1) it only uses a few scalars to parameterize the method and does not require the structural information of uncertainties; 2) it generalizes several classical filters (e.g., the fading Kalman filter, risk-sensitive Kalman filter, relative-entropy Kalman filter, outlier-insensitive Kalman filters) into a unified framework. We show that the distributionally robust state estimation problem can be reformulated into a linear semi-definite program and in some special cases it can be analytically solved. Comprehensive comparisons with existing state estimation frameworks that are insensitive to parameter uncertainties and measurement outliers are also conducted.
25 June 2020 in IEEE Internet of Things Journal
In order to lower the ramp risk and improve the aircraft ground handling efficiency, we aim to: (a) Track ground support equipment (GSE) in a real-time and high-accuracy manner so that we can not only conveniently obtain the positions and velocities of them but also reliably report latent collisions among aircraft and GSE. As a result, corresponding ramp risks could be detected and handled in advance; (b) Schedule the GSE in an optimal manner based on the real-time data gathered in advance to make efficient use of GSE so that we can smoothly serve the annually increasing air traffic while controlling the ramp area congestion and GSE overheads. In detail, firstly, we develop a real-time and high-accuracy tracking device consisting of one Real Time Kinematic (RTK) unit and Heading unit(s), for GSE including not only those which have only one carriage like tractors, shutters, and so forth, but also baggage transit trains which contain one tug plus multiple dollies. The tracking accuracy for GSE could be limited within centimeters so that the monitor, avoidance, and fixation of unaware ramp risks become possible. Secondly, for optimal scheduling of GSE, a mixed integer linear programming model and an efficient heuristic algorithm are proposed to minimize the total cost of equipment’s rental and travel consumption, while respecting the constraints like flights timetables, GSE moving speeds limit, the total number of GSE available in stock, and the maximum number of dollies allowed to attach to each baggage transit train, and so on.
5 June 2024 in IEEE Transactions on Signal Processing
Integrated sensing and communication (ISAC), which enables hardware, resources (e.g., spectra), and waveforms sharing, is becoming a key feature in future-generation communication systems. This paper investigates performance characterization and waveform design for ISAC systems when the underlying true communication channels are not accurately known. With uncertainty in a nominal communication channel, the nominal Pareto frontier of the sensing and communication performances cannot represent the true performance trade-off of a real-world operating ISAC system. Therefore, this paper portrays the robust (i.e., conservative) Pareto frontier considering the uncertainty in the communication channel. To be specific, the lower bound of the true (but unknown) Pareto frontier is investigated, technically by studying robust waveform design problems that find the best waveforms under the worst-case channels. The robust waveform design problems examined in this paper are shown to be non-convex and high-dimensional, which cannot be solved using existing optimization techniques. As such, we propose a computationally efficient solution framework to approximately solve them. Simulation results show that by solving the robust waveform design problems, the lower bound of the true but unknown Pareto frontier, which characterizes the sensing-communication performance trade-off under communication channel uncertainty, can be obtained.
7 October 2021 in IEEE Transactions on Signal Processing
Modeling uncertainties for real linear systems are unavoidable. These uncertainties can significantly degrade the performance of optimal state estimators designed for nominal system models. The challenge is quantifying such uncertainties and devising robust estimators that are insensitive to them. This paper is therefore concerned with distributionally robust state estimation for linear Markov systems. We propose a new modeling framework that describes uncertainties using a family of distributions so that the worst-case robust estimate in the state space is made over the least-favorable distribution. This framework uses only one or two scalars to express the uncertainty set and does not require the structural information of model uncertainties. Furthermore, the moment-based ambiguity set is suggested to embody the distributional uncertainty family. As a result, the estimation problem transforms into a nonlinear semidefinite program with linear constraints, which can be analytically and efficiently solved. Intensive experiments illustrate the advantages of the proposed framework over existing methods.
31 August 2022 in IEEE Transactions on Signal Processing
Uncertainties unavoidably exist in modeling for nonlinear systems: state equation, measurement equation, and/or noises statistics might be uncertain. Such model mismatches render the performance of nominally optimal state estimators being deteriorated or even unsatisfactory. Therefore, robust filters that are insensitive to modeling uncertainties have to be designed. The challenge is to quantitatively describe the uncertainties and then design accordingly efficient robust filters. Since uncertainties in nominal models make prior state distributions and likelihood distributions uncertain as well, this article proposes a distributionally robust particle filtering framework for nonlinear systems subject to modeling uncertainties. Specifically, we use worst-case prior state distributions (near the nominal prior state distributions) to generate prior state particles and/or determine their weights. Likewise, worst-case likelihood distributions (near the nominal likelihood distributions) are used to evaluate the worst-case likelihoods of prior state particles at given measurements. The “worst-case” scenario is quantified by entropy of distributions, and maximum entropy distributions are found in balls centered at nominal distributions with radii defined by statistical similarity measures such as moments-based similarity, Wasserstein distance, and Kullback-Leibler divergence. We prove that Gaussian approximation filters (e.g., unscented/cubature/ensemble Kalman filter) are distributionally robust in the sense that they use maximum entropy prior state distributions and maximum entropy likelihood distributions. Moreover, we show that the distributionally robust particle filtering framework provides a likelihood evaluation method for general nonlinear measurement equation with non-additive and non-multiplicative measurement noises. At last, we discuss measurement outlier treatment strategies in the distributionally robust particle filtering framework.
12 February 2021 in IEEE Transactions on Instrumentation and Measurement
Time series measurements from sensing units (e.g., UWB ranging circuits) always suffer from uncertainties like noises, outliers, dropouts, and/or nonspecific anomalies. In order to extract the true information with high precision from the original corrupted measurements, the signal-model-based signal pre-processing units embedded in sensing circuits are usually employed. However, for a general signal to observe, its signal model cannot be obtained so that the signal-model-based signal processing methods are not applicable. In this article, the time-variant local autocorrelated polynomial (TVLAP) model in the state space is proposed to model the dynamics of a non-stationary stochastic process (i.e., a signal or a time series), through which the model-based signal processing methods could be utilized to denoise, to correct the outliers/dropouts, and/or to identify anomalies contained in the measurements. Besides, the presented method can also be used in change point detection for a time series.