A noise estimator was presented in this paper by modeling the log-power sequence with hidden Markov model （HMM）. The smoothing factor of this estimator was motivated by the speech presence probability at each freque...A noise estimator was presented in this paper by modeling the log-power sequence with hidden Markov model （HMM）. The smoothing factor of this estimator was motivated by the speech presence probability at each frequency band. This HMM had a speech state and a nonspeech state, and each state consisted of a unique Gaussian function. The mean of the nonspeech state was the estimation of the noise logarithmic power. To make this estimator run in an on-line manner, an HMM parameter updated method was used based on a first-order recursive process. The noise signal was tracked together with the HMM to be sequentially updated. For the sake of reliability, some constraints were introduced to the HMM. The proposed algorithm was compared with the conventional ones such as minimum statistics （MS） and improved minima controlled recursive averaging （IM- CRA）. The experimental results confirms its promising performance.展开更多
The output feedback model predictive control(MPC),for a linear parameter varying(LPV) process system including unmeasurable model parameters and disturbance(all lying in known polytopes),is considered.Some previously ...The output feedback model predictive control(MPC),for a linear parameter varying(LPV) process system including unmeasurable model parameters and disturbance(all lying in known polytopes),is considered.Some previously developed tools,including the norm-bounding technique for relaxing the disturbance-related constraint handling,the dynamic output feedback law,the notion of quadratic boundedness for specifying the closed-loop stability,and the ellipsoidal state estimation error bound for guaranteeing the recursive feasibility,are merged in the control design.Some previous approaches are shown to be the special cases.An example of continuous stirred tank reactor(CSTR) is given to show the effectiveness of the proposed approaches.展开更多
This paper addresses the state estimation problem for linear systems with additive uncertainties in both the state and output equations using a moving horizon approach. Based on the full information estimation setting...This paper addresses the state estimation problem for linear systems with additive uncertainties in both the state and output equations using a moving horizon approach. Based on the full information estimation setting and the game-theoretic approach to the H∞filtering, a new optimization-based estimation scheme for uncertain linear systems is proposed, namely the H∞-full information estimator, H∞-FIE in short. In this formulation, the set of processed data grows with time as more measurements are received preventing recursive formulations as in Kalman filtering. To overcome the latter problem, a moving horizon approximation to the H∞-FIE is also presented, the H∞-MHE in short. This moving horizon approximation is achieved since the arrival cost is suitably defined for the proposed scheme. Sufficient conditions for the stability of the H∞-MHE are derived. Simulation results show the benefits of the proposed scheme when compared with two H∞filters and the well-known Kalman filter.展开更多
基金Supported by the National Key Basic Research Program of China(2013CB329302)the National Natural Science Foundation of China(61271426,10925419,90920302,61072124,11074275,11161140319,91120001)+3 种基金the Strategic Priority Research Program of the Chinese Academy of Sciences(XDA06030100,XDA06030500)the National "863" Program(2012AA012503)the CAS Priority Deployment Project(KGZD-EW-103-2)Jiangxi Provincial Department of Education Science and Technology Project(GJJ13426)
文摘A noise estimator was presented in this paper by modeling the log-power sequence with hidden Markov model （HMM）. The smoothing factor of this estimator was motivated by the speech presence probability at each frequency band. This HMM had a speech state and a nonspeech state, and each state consisted of a unique Gaussian function. The mean of the nonspeech state was the estimation of the noise logarithmic power. To make this estimator run in an on-line manner, an HMM parameter updated method was used based on a first-order recursive process. The noise signal was tracked together with the HMM to be sequentially updated. For the sake of reliability, some constraints were introduced to the HMM. The proposed algorithm was compared with the conventional ones such as minimum statistics （MS） and improved minima controlled recursive averaging （IM- CRA）. The experimental results confirms its promising performance.
基金Supported by the National High Technology Research and Development Program of China(2014AA041802)the National Natural Science Foundation of China(61573269)
文摘The output feedback model predictive control(MPC),for a linear parameter varying(LPV) process system including unmeasurable model parameters and disturbance(all lying in known polytopes),is considered.Some previously developed tools,including the norm-bounding technique for relaxing the disturbance-related constraint handling,the dynamic output feedback law,the notion of quadratic boundedness for specifying the closed-loop stability,and the ellipsoidal state estimation error bound for guaranteeing the recursive feasibility,are merged in the control design.Some previous approaches are shown to be the special cases.An example of continuous stirred tank reactor(CSTR) is given to show the effectiveness of the proposed approaches.
基金supported by the European Community s Seventh Framework Programme FP7/2007-2013(No.223854)COLCIENCIAS-Departamento Administrativo de Ciencia,Tecnologíae Innovacin de Colombia
文摘This paper addresses the state estimation problem for linear systems with additive uncertainties in both the state and output equations using a moving horizon approach. Based on the full information estimation setting and the game-theoretic approach to the H∞filtering, a new optimization-based estimation scheme for uncertain linear systems is proposed, namely the H∞-full information estimator, H∞-FIE in short. In this formulation, the set of processed data grows with time as more measurements are received preventing recursive formulations as in Kalman filtering. To overcome the latter problem, a moving horizon approximation to the H∞-FIE is also presented, the H∞-MHE in short. This moving horizon approximation is achieved since the arrival cost is suitably defined for the proposed scheme. Sufficient conditions for the stability of the H∞-MHE are derived. Simulation results show the benefits of the proposed scheme when compared with two H∞filters and the well-known Kalman filter.
