检索结果-南通市图书馆

A new self-learning optimal control laws for a class of discrete-time nonlinear systems based on ESN architecture

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2014年第6期57卷 284-293页

作者： SONG RuiZhuo xiao wendong SUN ChangYin School of Automation and Electrical Engineering University of Science and Technology Beijing

A novel self-learning optimal control method for a class of discrete-time nonlinear systems is proposed based on iteration adaptive dynamic programming(ADP)*** is proven that the iteration costate functions converge to the optimal one,and a detailed convergence analysis of the iteration ADP algorithm is ***,echo state network(ESN)architecture is used as the approximator of the costate function for each *** ensure the reliability of the ESN approximator,the ESN mean square training error is constrained in the satisfactory *** simulation examples are given to demonstrate that the proposed control method has a fast response speed due to the special structure and the fast training process.

关键词： adaptive dynamic programming discrete-time optimal control ESN costate function

Sensor selection for parameterized random field estimation in wireless sensor networks

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

控制理论与应用（英文版） 2011年第1期9卷 44-50页

作者： Weng, Yang xiao, wendong Xie, Lihua School of Electrical and Electronic Engineering Nanyang Technological University Singapore 639798 Singapore Institute for Infocomm Research Singapore 138632 Singapore School of Mathematics Sichuan University Chengdu Sichuan 610064 China

We consider the random field estimation problem with parametric trend in wireless sensor networks where the field can be described by unknown parameters to be estimated. Due to the limited resources, the network selects only a subset of the sensors to perform the estimation task with a desired performance under the D-optimal criterion. We propose a greedy sampling scheme to select the sensor nodes according to the information gain of the sensors. A distributed algorithm is also developed by consensus-based incremental sensor node selection through information quality computation for and message exchange among neighboring sensors. Simulation results show the good performance of the proposed algorithms.

关键词： Random field estimation Parametric trend Wireless sensor network Sensor selection NP-completeness Distributed processing

Optimal full-order filtering for discrete-time systems with random measurement delays and multiple packet dropouts

在线全文

学校读者我要写书评

暂无评论

控制理论与应用（英文版） 2010年第1期8卷 105-110页

作者： Shuli SUN Lihua XIE wendong xiao School of Electronic Engineering Heilongjiang University Harbin Heilongjiang 150080 China School of Electrical and Electronic Engineering Nanyang Technological University Singapore 639798 Institute for Infocomm Research Singapore 138632

This paper is concerned with the estimation problem for discrete-time stochastic linear systems with possible single unit delay and multiple packet dropouts. Based on a proposed uncertain model in data transmission, an optimal full-order filter for the state of the system is presented, which is shown to be of the form of employing the received outputs at the current and last time instants. The solution to the optimal filter is given in terms of a Riccati difference equation governed by two binary random variables. The optimal filter is reduced to the standard Kalman filter when there are no random delays and packet dropouts. The steady-state filter is also investigated. A sufficient condition for the existence of the steady-state filter is given. The asymptotic stability of the optimal filter is analyzed.

关键词： Full-order filter Random delay Packet dropouts Riccati difference equation

Sensor selection for received signal strength-based source localization in wireless sensor networks

在线全文

学校读者我要写书评

暂无评论

控制理论与应用（英文版） 2011年第1期9卷 51-57页

作者： Zhao, Bin Guan, Xinping Xie, Lihua xiao, wendong Key Laboratory of Industrial Computer Control Engineering of Hebei Province Institute of Electrical Engineering Yanshan University Qinhuangdao Hebei 066004 China School of Electronics Information and Electrical Shanghai Jiao Tong University Shanghai 200240 China School of Electrical and Electronic Engineering Nanyang Technological University Singapore 639798 Singapore Institute for Infocomm Research Singapore 138632 Singapore

Generally, localization is a nonlinear problem, while linearization is used to simplify this problem. Reasonable approximations could be achieved when signal-to-noise ratio (SNR) is large enough. Energy is a critical resource in wireless sensor networks, and system lifetime needs to be prolonged through the use of energy efficient strategies during system operation. In this paper, a closed-form solution for received signal strength (RSS)-based source localization in wireless sensor network (WSN) is obtained. A sensor selection method is proposed to improve the localization accuracy as well as to save energy consumption. By selecting only a limited number of sensor nodes based on the model accuracy and geometry structure analysis, localization performance is improved, and energy consumption is reduced. In addition, extensive simulations are presented to demonstrate that the estimation performance with the proposed sensor selection method is better than that without sensor selection.

关键词： Received signal strength Source localization Closed-form solution Sensor selection Geometry structure

Energy-efficient adaptive sensor scheduling for target tracking in wireless sensor networks

在线全文

学校读者我要写书评

暂无评论

控制理论与应用（英文版） 2010年第1期8卷 86-92页

作者： wendong xiao Sen ZHANG Jianyong LIN Chen Khong THAM Institute for lnfocomm Research 1 Fusionopolis Way #21-01 Connexis Singapore 138632 School of Mechanical and Aeronautical Engineering Singapore Polytechnic 500 Dover Road Singapore 139651 School of Electrical and Electronic Engineering Nanyang Technological University Singapore 639798

Sensor scheduling is essential to collaborative target tracking in wireless sensor networks （WSNs）. In the existing works for target tracking in WSNs, such as the information-driven sensor query （IDSQ）, the tasking sensors are scheduled to maximize the information gain while minimizing the resource cost based on the uniform sampling intervals, ignoring the changing of the target dynamics and the specific desirable tracking goals. This paper proposes a novel energy-efficient adaptive sensor scheduling approach that jointly selects tasking sensors and determines their associated sampling intervals according to the predicted tracking accuracy and tracking energy cost. At each time step, the sensors are scheduled in alternative tracking mode, namely, the fast tracking mode with smallest sampling interval or the tracking maintenance mode with larger sampling interval, according to a specified tracking error threshold. The approach employs an extended Kalman filter （EKF）-based estimation technique to predict the tracking accuracy and adopts an energy consumption model to predict the energy cost. Simulation results demonstrate that, compared to a non-adaptive sensor scheduling approach, the proposed approach can save energy cost significantly without degrading the tracking accuracy.

关键词： Wireless sensor network Target tracking Sensor scheduling Extended Kalman filter Energy efficiency.

Large-Scale WiFi Indoor Localization via Extreme Learning Machine 36

在线全文

学校读者我要写书评

暂无评论

Large-Scale WiFi Indoor Localization via Extreme Learning Ma...

第36届中国控制会议

作者： Jie Zhang Jian Sun Hailong Wang wendong xiao Lin Tan School of Automation & Electrical Engineering University of Science and Technology Beijing School of Mechanical Engineering & Automation Beihang University The First Research Institute of Ministry of Public Security

ISBN: (纸本)9781538629185

Due to the widespread deployment and low cost,WiFi has become one of the most attractive research fields in indoor environments,for example,almost every smart mobile device can be a WiFi signal receiver and *** WiFi based indoor localization approaches have been proposed under different ***,most of them are only suitable for relatively small indoor environments,and when we implement these approaches to large-scale environments,they may encounter the scalability problem due to the huge received signal strength（RSS） database,and also suffer from time-consuming,poor accuracy,*** order to tackle these issues,we propose a modified indoor localization scheme for large-scale area by utilizing extreme learning machine（ELM）and K-means *** whole localization process involves two phases,i.e.,the offline phase and the online *** the offline phase,we firstly divide the monitoring area into some smaller parts by K-means clustering,and then an ELM classifier will be trained for identifying the divided parts,after that ELMs for every divided part will be trained using the collected received signal strength（RSS） *** the online phase,the attribution of the localization point is firstly judged by the ELM classifier,and then the localization of the target will be estimated through the corresponding *** performance of the proposed approach is verified in both small indoor area,and larger-scale more complicated indoor *** results show that the proposed scheme can obtain better localization accuracy than existing schemes.

关键词： Large-scale Indoor Localization Received Signal Strength(RSS) Extreme Learning Machine(ELM) K-means Clustering

A Weighted Voting Ensemble Prediction Model For Adjusting Burden Distribution Matrix Of Blast Furnace

在线全文

cnki会议

学校读者我要写书评

暂无评论

A Weighted Voting Ensemble Prediction Model For Adjusting Bu...

第40届中国控制会议

作者： Jiangxiao Wu Sen Zhang wendong xiao School of Automation University of Science and Technology Beijing Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) University of Science and Technology Beijing

Burden distribution is an important method to regulate the operation state of blast furnace. Reasonable burden distribution matrix of blast furnace is the prerequisite for improving resource utilization and ensuring stable blast furnace *** order to achieve the above target, an ensemble prediction model based on weighted voting and a method of calculating weights are proposed in this paper. In this model, the convolution neural network（CNN）-bidirectional gated recurrent unit（BiGRU）based on attention mechanism, logistic regression, k-nearest neighbor（KNN） and support vector machines（SVM） are integrated to predict whether the burden distribution matrix of blast furnace needs to be adjusted. The weight of each individual model is determined by its prediction accuracy and F1-score. CNN converts the extracted feature vectors into time series and inputs them into BiGRU. The attention mechanism is introduced to assign different weights to the hidden state of BiGRU, so as to enhance the long-term influence of important information. Finally, the production data of blast furnace are used to testify the performance of prediction models. In this paper, Multilayer Perceptron（MLP） model, RNN model, BiGRU model, BiGRU model based on attention mechanism, logistic regression model, KNN model, SVM model and the proposed ensemble model are used for comparison experiments, compared with other prediction models, the experimental results illustrated that the weighted voting ensemble prediction model can accurately determine whether the burden distribution matrix of blast furnace needs to be adjusted and have better prediction accuracy and generalization performance.

关键词： Blast furnace Burden distribution matrix Ensemble model Attention mechanism BiGRU

Cost-sensitive multi-layer perceptron for binary classification with imbalanced data

在线全文

cnki会议

学校读者我要写书评

暂无评论

Cost-sensitive multi-layer perceptron for binary classificat...

第37届中国控制会议

作者： Zheng Liu Sen Zhang wendong xiao Yan Di School of Automation and Electrical Engineering University of Science and Technology Beijing Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education School of Automation and Electrical EngineeringUniversity of Science and Technology Beijing

Currently, class imbalance has been a challenge for classification due to its highly imbalanced instances of distinct classes. With the advantage in quantity, the majority classes can get high accuracy in classification while many instances belonging to minority classes are inclined to be classified as majority classes. In this paper, we propose a novel cost-sensitive method based on multi-layer perceptron（CMMLP） for binary classification with imbalanced data. The proposed cost matrix is used to modify the construction of loss function so as to encourage classifier to pay more attention to the accuracy of minority class by minimizing training error. In order to verify the effectiveness of CMMLP, CMMLP is applied to some benchmark datasets and Click-Through Rate（CTR） prediction datasets. Experimental results illustrate that the new cost-sensitive approach can achieve better performance for binary classification with imbalanced data than the original MLP（OMLP）.

关键词： Imbalance learning Binary classification CMMLP CTR prediction

Device-Free Localization Using Empirical Wavelet Transform-based Extreme Learning Machine

在线全文

cnki会议

学校读者我要写书评

暂无评论

Device-Free Localization Using Empirical Wavelet Transform-b...

第30届中国控制与决策会议

作者： Jie Zhang Yifang Lu Baoqiang Zhang wendong xiao School of Automation & Electrical Engineering University of Science and Technology Beijing Beijing Engineering Research Center of Industrial Spectrum Imaging State Grid Jincheng Power Supply Company

Device-Free Localization（DFL） is becoming one of the new trends in the localization field and has attracted more attentions, due to its advantage that the target to be localized does not need to be attached to any electronic *** our previous work, we have applied extreme learning machine（ELM） to DFL combining parameterized geometrical feature extraction（PGFE） for extending the input feature of ELM. Although, the proposed PGFE-ELM for DFL is more robust and can obtain better performance, we believe that the localization accuracy of ELM-based DFL can be improved further, due to the dynamic and uncertain signal propagation environment in the monitoring area. Thus, we will propose a novel ELM based on empirical wavelet transform（EWT）, named EWT-ELM, to improve the accuracy and efficiency of DFL. EWT-ELM consists of two components, i.e., the prediction model and the residual model. The prediction model is used for building the feature mapping of the input and the output through an ELM. In the residual model, the residual of the prediction model is firstly decomposed into some modes by EWT for alleviating the negative impacts of its nonlinear characteristic, and then, the decomposed modes will be modeled through the corresponding ELMs. In addition, the proposed EWT-ELM for DFL is trained in the offline training phase and performed for real-time estimation of target′s location in the online localization phase. Experimental results show that the proposed EWT-ELM can improve the localization accuracy significantly comparing with the classic ELM.

关键词： Device-Free Localization Extreme Learning Machine Empirical Wavelet Transform