Current Issue

11 October 2010, Volume 21 Issue 5
Hybrid pilots assisted channel estimation algorithm for MIMO-OFDM systems
Yang Zhang, Jiandong Li, and Lihua Pang
2010, 21(5):  721-728.  doi:10.3969/j.issn.1004-4132.2010.05.001
Abstract ( )   PDF (931KB) ( )  
Related Articles | Metrics

Abstract: A hybrid pilots assisted channel estimation algorithm for multiple input multiple output (MIMO) orthogonal frequency division multiplexing (OFDM) systems under low signal-to-noise ratio (SNR) and arbitrary Doppler spread scenarios is proposed. Motivated by the dissatisfactory performance of the optimal pilots (OPs) designed under static channels over multiple OFDM symbols imposed by fast fading channels, the proposed scheme first assumes that the virtual pilot tones superimposed at data locations over specific subcarriers are transmitted from all antennas, then the virtual received pilot signals at the corresponding locations can be obtained by making full use of the time and frequency domain correlations of the frequency responses of the time varying dispersive fading channels and the received signals at pilot subcarriers, finally the channel parameters are derived from the combination of the real and virtual received pilot signals over one OFDM symbol based on least square (LS) criterion. Simulation results illustrate that the proposed method is insensitive to Doppler spread and can effectively ameliorate the mean square error (MSE) floor inherent to the previous method, meanwhile its performance outmatches that of OPs at low SNR region under static channels.

Low-complexity method for DOA estimation based on ESPRIT
Xuebin Zhuang, Xiaowei Cui, Mingquan Lu, and Zhenming Feng
2010, 21(5):  729-733.  doi:10.3969/j.issn.1004-4132.2010.05.002
Abstract ( )   PDF (652KB) ( )  
Related Articles | Metrics

A low-complexity method for direction of arrival (DOA) estimation based on estimation signal parameters via rotational invariance technique (ESPRIT) is proposed. Instead of using the cross-correlation vectors in multistage Wiener filter (MSWF), the orthogonal residual vectors obtained in conjugate gradient (CG) method span the signal subspace used by ESPRIT. The computational complexity of the proposed method is significantly reduced, since the signal subspace estimation mainly needs two matrixvector complex multiplications at the iteration of data level. Furthermore, the prior training data are not needed in the proposed
method. To overcome performance degradation at low signal to noise ratio (SNR), the expanded signal subspace spanned by more basis vectors is used and simultaneously renders ESPRIT yield redundant DOAs, which can be excluded by performing ESPRIT once more using the unexpanded signal subspace. Compared
with the traditional ESPRIT methods by MSWF and eigenvalue decomposition (EVD), numerical results demonstrate the satisfactory performance of the proposed method.

LLR calculation for LDPC coded SCBT in 60 GHz WPAN
Lingjun Kong,Yang Xiao,Ming Lei, and Ye Huang
2010, 21(5):  734-739.  doi:10.3969/j.issn.1004-4132.2010.05.003
Abstract ( )   PDF (2599KB) ( )  
Related Articles | Metrics

The single-carrier block transmission (SCBT), a.k.a., single-carrier frequency-domain equalization (SC-FDE), is being considered as an option technique for the wireless personal area network (WPAN) operating at 60 GHz. It is found that for residential environment, in non-line-of-sight (NLOS) multi-path channels, the SCBT is much more effective to combat the inter-symbol interference (ISI) compared with orthogonal frequency division multiplexing (OFDM). Low-density parity-check (LDPC) codes are a class of linear block codes which provide near capacity performance on a large collection of data transmission and storage channels while simultaneously admitting implementable decoders. To facilitate using LDPC codes for SCBT system, a new log-likelihood ratio (LLR) calculation method is proposed based on pilot symbols (PS). Golay Sequences whose sum autocorrelation has a unique peak and zero sidelobe are used for creating the PS. The position and length of the PS are not fixed in the data blocks. The simulation results show that the proposed method can significantly improve the LDPC decoding performance in SCBT system. This is very promising to support ultra high-data-rate wireless transmission.

Moving object detection in framework of compressive sampling
Jing Li, Junzheng Wang, and Wei Shen
2010, 21(5):  740-745.  doi:10.3969/j.issn.1004-4132.2010.05.004
Abstract ( )   PDF (866KB) ( )  
Related Articles | Metrics

Compressive sensing is a revolutionary idea proposed recently to achieve much lower sampling rate for signals. In the image application with limited resources the camera data can be stored and processed in compressed form. An algorithm for moving object and region detection in video using a compressive sampling is developed. The algorithm estimates motion information of the moving object and regions in the video from the compressive measurements of the current image and background scene. The algorithm does not perform inverse compressive operation to obtain the actual pixels of the current image nor the estimated background. This leads to a computationally efficient method and a system compared with the existing motion estimation methods. The experimental results show that the sampling rate can reduce to 25% without sacrificing performance.

Velocity compensation methods for LPRF modulated frequency stepped-frequency (MFSF) radar
Guifen Xia, Hongyan Su, and Peikang Huang
2010, 21(5):  746-751.  doi:10.3969/j.issn.1004-4132.2010.05.005
Abstract ( )   PDF (2069KB) ( )  
Related Articles | Metrics

In the high speed target environment, there exists serious Doppler effect in the low pulse repetition frequency (LPRF) modulated frequency stepped frequency (MFSF) radar signal. The velocity range of the target is large and the velocity is high ambiguous, so the single method is difficult to satisfy the velocity measurement requirement. For this problem, a novel method is presented, it is a combination of cross-correlation inner frame velocity measurement and range-Doppler coupling velocity measurement. The cross-correlation inner frame method, overcoming the low Doppler tolerance of the cross-correlation between frames, can obtain the coarse velocity of the high speed target, and then the precision velocity can be obtained with the range-Doppler coupling method. The simulation results confirm the method is effective, and also it is well real-time and easy to the project application.

Target location with signal fitting and sub-aperture tracking for airborne multi-channel radar
Zhiwei Yang, Guisheng Liao, Shun He, and Cao Zeng
2010, 21(5):  752-758.  doi:10.3969/j.issn.1004-4132.2010.05.006
Abstract ( )   PDF (1433KB) ( )  
Related Articles | Metrics

The location of a moving target based on signal fitting and sub-aperture tracking from an airborne multi-channel radar is dealt with. The proposed approach is applied in two steps: first, the ambiguous slant-range velocity is derived with a modified single-snapshot multiple direction of arrival estimation method, and second, the unambiguous slant-range velocity is found using a track-based criterion. The prominent advantage of the proposed approach is that the unambiguous slant-range velocity can be very large. Besides, the first stage is carried out at the determinate range-Doppler test cell by azimuth searching for fitting best to the moving target signal, therefore, the location performance would not be sacrificed in order to suppress clutter and/or interference. The effectiveness and efficiency of the proposed method are validated with a set of airborne experimental data.

Novel matched filtering method and its application
Fei Meng, Lianggui Xie, and Raohui Li
2010, 21(5):  759-762.  doi:10.3969/j.issn.1004-4132.2010.05.007
Abstract ( )   PDF (863KB) ( )  
Related Articles | Metrics
The method of using a narrowband filter to realize matched filtering is derived. A novel method of using spectrum sampling to realize matched filtering is presented, and the method can conquer the disadvantages that the narrowband filter cannot adopt the adaptive scheduling of phased array radars and realize matched filtering for several waveforms. A novel error extraction method is proposed, which uses a time division multipath method to realize the intermediate frequency extraction. This method can save lots of space for vehicle-born radar systems, reduce the influence of amplitude and phase distortion caused by devices, and enhance the system reliability. Simulation results show that the spectrum sampling method is applicable, and the implementation of frequency spectrum sampling is elaborated.
Orthogonal genetic algorithm for solving quadratic bilevel programming problems
Hong Li, Yongchang Jiao, and Li Zhang
2010, 21(5):  763-770.  doi:10.3969/j.issn.1004-4132.2010.05.008
Abstract ( )   PDF (403KB) ( )  
Related Articles | Metrics

A quadratic bilevel programming problem is transformed  into a single level complementarity slackness problem by applying  Karush-Kuhn-Tucker (KKT) conditions. To cope with the complementarity  constraints, a binary encoding scheme is adopted for  KKT multipliers, and then the complementarity slackness problem  is simplified to successive quadratic programming problems,  which can be solved by many algorithms available. Based on 01  binary encoding, an orthogonal genetic algorithm, in which the orthogonal  experimental design with both two-level orthogonal array  and factor analysis is used as crossover operator, is proposed.  Numerical experiments on 10 benchmark examples show that the  orthogonal genetic algorithm can find global optimal solutions of  quadratic bilevel programming problems with high accuracy in a  small number of iterations.

Integrating dual-role variables in data envelopment analysis
Feng Yang, Liang Liang, Zhaoqiong Li, and Shaofu Du
2010, 21(5):  771-776.  doi:10.3969/j.issn.1004-4132.2010.05.009
Abstract ( )   PDF (505KB) ( )  
Related Articles | Metrics

Traditional data envelopment analysis (DEA) theory assumes that decision variables are regarded as inputs or outputs, and no variable can play the roles of both an input and an output at the same time. In fact, there exist some variables that work as inputs and outputs simultaneously and are called dual-role variables. Traditional DEA models cannot be used to appraise the performance of decision making units containing dual-role variables. The paper analyzes the structure and properties of the production systems comprising dual-role variables, and proposes a DEA model integrating dual-role variables. Finally the proposed model is illustrated to evaluate the efficiency of university departments.  

Multi-population and diffusion UMDA for dynamic multimodal problems
Yan Wu, Yuping Wang, Xiaoxiong Liu, and Jimin Ye
2010, 21(5):  777-783.  doi:10.3969/j.issn.1004-4132.2010.05.010
Abstract ( )   PDF (393KB) ( )  
Related Articles | Metrics
In dynamic environments, it is important to track changing optimal solutions over time. Univariate marginal distribution algorithm (UMDA) which is a class algorithm of estimation of distribution algorithms attracts more and more attention in recent years. In this paper a new multi-population and diffusion UMDA (MDUMDA) is proposed for dynamic multimodal problems. The multi-population approach is used to locate multiple local optima which are useful to find the global optimal solution quickly to dynamic multimodal problems. The diffusion model is used to increase the diversity in a guided fashion, which makes the neighbor individuals of previous optimal solutions move gradually from the previous optimal solutions and enlarge the search space. This approach uses both the information of current population and the part history information of the optimal solutions. Finally experimental studies on the moving peaks benchmark are carried out to evaluate the proposed algorithm and compare the performance of MDUMDA and multi-population quantum swarm optimization (MQSO) from the literature. The experimental results show that the MDUMDA is effective for the function with moving optimum and can adapt to the dynamic environments rapidly.   
Analysis and verification of network profile
Weiwei Chen, Ning Huang, Yuqing Liu, Ye Wang, and Rui Kang
2010, 21(5):  784-790.  doi:10.3969/j.issn.1004-4132.2010.05.011
Abstract ( )   PDF (1026KB) ( )  
Related Articles | Metrics

The elements of network profile are proposed. Based   on the network traffic distribution model, the network profile includes   the application request rate, the branch transfer probability,   the ratio of application requests, and the probability distribution of   the requested objects. Based on the evaluation method of network   performance reliability, four simulation cases are constructed in   OPNET software, and the results show the four elements of profile   have impacts on the network reliability.

Multiobjective maintenance optimization of the continuously monitored deterioration system
Changyou Li, Minqiang Xu, Song Guo, and Rixin Wang
2010, 21(5):  791-795.  doi:10.3969/j.issn.1004-4132.2010.05.012
Abstract ( )   PDF (1065KB) ( )  
Related Articles | Metrics

With the development of the monitoring technology, it is more and more common that the system is continuously monitored. Therefore, the research on the maintenance optimization of the continuously monitored deterioration system is important. The deterioration process of the discussed system is described by a Gamma process. The predictive maintenance is considered to be imperfect and formulated. The expected interval of two continuous preventive maintenances is derived. Then, the maintenance optimization model of the continuously monitored deterioration system is presented. In the model, the minimization of the expected operational cost per unit time and the maximization of the system availability are the optimization objectives. The improved ideal point method with the normalized objective functions is employed to solve the proposed model. The validity and sensitivity of the proposed multiobjective maintenance optimization model are analyzed by a numerical example.  

Improved differential evolution algorithm for resource-constrained project scheduling problem
Lianghong Wu, Yaonan Wang, and Shaowu Zhou
2010, 21(5):  798-805.  doi:10.3969/j.issn.1004-4132.2010.05.013
Abstract ( )   PDF (1337KB) ( )  
Related Articles | Metrics

An improved differential evolution (IDE) algorithm that adopts a novel mutation strategy to speed up the convergence rate is introduced to solve the resource-constrained project scheduling problem (RCPSP) with the objective of minimizing project duration. Activities priorities for scheduling are represented by individual vectors and a serial scheme is utilized to transform the individual-represented priorities to a feasible schedule according to the precedence and resource constraints so as to be evaluated. To investigate the performance of the IDE-based approach for the RCPSP, it is compared against the meta-heuristic methods of hybrid genetic algorithm (HGA), particle swarm optimization (PSO) and several well selected heuristics. The results show that the proposed scheduling method is better than general heuristic rules and is able to obtain the same optimal result as the HGA and PSO approaches but more efficient than the two algorithms.  

Solving DCLAP-MSN based on hybrid genetic algorithm
Hongtao Lei, Bo Guo, and Tao Zhang
2010, 21(5):  806-811.  doi:10.3969/j.issn.1004-4132.2010.05.014
Abstract ( )   PDF (1139KB) ( )  
Related Articles | Metrics
Modeling of combined Bayesian networks and cognitive framework for decision-making in C2
Li Wang, and Mingzhe Wang
2010, 21(5):  812-820.  doi:10.3969/j.issn.1004-4132.2010.05.015
Abstract ( )   PDF (2440KB) ( )  
Related Articles | Metrics

The command and control (C2) is a decision-making process based on human cognition, which contains operational, physical, and human characteristics, so it takes on uncertainty and complexity. As a decision support approach, Bayesian networks (BNs) provide a framework in which a decision is made by combining the experts’ knowledge and the specific data. In addition, an expert system represented by human cognitive framework is adopted to express the real-time decision-making process of the decision maker. The combination of the Bayesian decision support and human cognitive framework in the C2 of a specific application field is modeled and executed by colored Petri nets (CPNs), and the consequences of execution manifest such combination can perfectly present the decision-making process in C2.  

Modeling for UAV resource scheduling under mission synchronization
Jia Zeng, Xiaoke Yang, Lingyu Yang, and Gongzhang Shen
2010, 21(5):  821-826.  doi:10.3969/j.issn.1004-4132.2010.05.016
Abstract ( )   PDF (451KB) ( )  
Related Articles | Metrics

Unmanned aerial vehicle (UAV) resource scheduling means to allocate and aggregate the available UAV resources depending on the mission requirements and the battlefield situation assessment. In previous studies, the models cannot reflect the mission synchronization; the targets are treated respectively, which results in the large scale of the problem and high computational complexity. To overcome these disadvantages, a model for UAV resource scheduling under mission synchronization is proposed, which is based on single-objective non-linear integer programming. And several cooperative teams are aggregated for the target clusters from the available resources. The evaluation indices of weapon allocation are referenced in establishing the objective function and the constraints for the issue. The scales of the target clusters are considered as the constraints for the scales of the cooperative teams to make them match in scale. The functions of the intersection between the “mission time-window” and the UAV “arrival time-window” are introduced into the objective function and the constraints in order to describe the mission synchronization effectively. The results demonstrate that the proposed expanded model can meet the requirement of mission synchronization, guide the aggregation of cooperative teams for the target clusters and control the scale of the problem effectively.  

Air-combat behavior data mining based on truncation method
Yunfei Yin, Guanghong Gong, and Liang Han
2010, 21(5):  827-832.  doi:10.3969/j.issn.1004-4132.2010.05.017
Abstract ( )   PDF (1069KB) ( )  
Related Articles | Metrics

This paper considers the problem of applying data mining techniques to aeronautical field. The truncation method, which is one of the techniques in the aeronautical data mining, can be used to efficiently handle the air-combat behavior data. The technique of air-combat behavior data mining based on the truncation method is proposed to discover the air-combat rules or patterns. The simulation platform of the air-combat behavior data mining that supports two fighters is implemented. The simulation experimental results show that the proposed air-combat behavior data mining technique based on the truncation method is feasible whether in efficiency or in effectiveness.

H∞ fault estimation for a class of linear time-delay systems in finite frequency domain
Quanchao Dong, Maiying Zhong, and Steven X. Ding
2010, 21(5):  835-841.  doi:10.3969/j.issn.1004-4132.2010.05.018
Abstract ( )   PDF (472KB) ( )  
Related Articles | Metrics

This paper deals with the problem of Hfault estimation for linear time-delay systems in finite frequency domain. First a generalized coordinate change is applied to the original system such that in the new coordinates all the time-delay terms are injected by the system’s input and output. Then an observer-based Hfault estimator with input and output injections is proposed for fault estimation with known frequency range. With the aid of Generalized Kalman-Yakubovich-Popov lemma, sufficient conditions on the existence of the Hfault estimator are derived and a solution to the observer gain matrices is obtained by solving a set of linear matrix inequalities. Finally, a numerical example is given to illustrate the effectiveness of the proposed method.

Robust decentralized adaptive output feedback fuzzy controller design and application to AHS
Yishao Huang, Dequn Zhou, Xiaoxin Chen, Xueyun Chen, and Zhengwu Wang
2010, 21(5):  842-849.  doi:10.3969/j.issn.1004-4132.2010.05.019
Abstract ( )   PDF (815KB) ( )  
Related Articles | Metrics

A novel decentralized indirect adaptive output feedback fuzzy controller is developed for a class of large-scale uncertain nonlinear systems using error filtering. By the properly filtering of the observation error dynamics, the strictly positive-real condition is guaranteed to hold such that the proposed output feedback and adaptation mechanisms are practicable in practice owing to the fact that its implementation does not require the observation error vector itself any more, which corrects the impracticable schemes in the previous literature involved. The presented control algorithm can ensure that all the signals of the closed-loop large-scale system keep uniformly ultimately bounded and that the tracking error converges to zero asymptotically. The decentralized output feedback fuzzy controller can be applied to address the longitudinal control problem of a string of vehicles within an automated highway system (AHS) and the effectiveness of the design procedure is supported by simulation results.  

Adaptive output-feedback control for MIMO nonlinear systems with time-varying delays using neural networks
Weisheng Chen and Ruihong Li
2010, 21(5):  850-858.  doi:10.3969/j.issn.1004-4132.2010.05.020
Abstract ( )   PDF (604KB) ( )  
Related Articles | Metrics

An adaptive neural network output-feedback regulation approach is proposed for a class of multi-input-multi-output nonlinear time-varying delayed systems. Both the designed observer and controller are free from time delays. Different from the existing results, this paper need not the assumption that the upper bounding functions of time-delay terms are known, and only a neural network is employed to compensate for all the upper bounding functions of time-delay terms, so the designed controller procedure is more simplified. In addition, the resulting closed-loop system is proved to be semi-globally ultimately uniformly bounded, and the output regulation error converges to a small residual set around the origin. Two simulation examples are provided to verify the effectiveness of control scheme.

Constrained predictive control of nonlinear stochastic systems
Yanyan Yin and Fei Liu
2010, 21(5):  859-867.  doi:10.3969/j.issn.1004-4132.2010.05.021
Abstract ( )   PDF (242KB) ( )  
Related Articles | Metrics

The receding horizon control (RHC) problem is considered for nonlinear Markov jump systems which can be represented by Takagi-Sugeno fuzzy models subject to constraints both on control inputs and on observe outputs. In the given receding horizon, for each mode sequence of the T-S modeled nonlinear system with Markov jump parameter, the cost function is optimized by constraints on state trajectories, so that the optimization control input sequences are obtained in order to make the state into a terminal invariant set. Out of the receding horizon, the stability is guaranteed by searching a state feedback control law. Based on such stability analysis, a linear matrix inequality approach for designing receding horizon predictive controller for nonlinear systems subject to constraints both on the inputs and on the outputs is developed. The simulation shows the validity of this method.    

Adaptive functional link network control of near-space vehicles with dynamical uncertainties
Yanli Du, Qingxian Wu, Changsheng Jiang, and Jie Wen
2010, 21(5):  868-876.  doi:10.3969/j.issn.1004-4132.2010.05.022
Abstract ( )   PDF (1980KB) ( )  
Related Articles | Metrics

The control law design for a near-space hypersonic vehicle (NHV) is highly challenging due to its inherent nonlinearity, plant uncertainties and sensitivity to disturbances. This paper presents a novel functional link network (FLN) control method for an NHV with dynamical thrust and parameter uncertainties. The approach devises a new partially-feedback-functional-link-network (PFFLN) adaptive law and combines it with the nonlinear generalized predictive control (NGPC) algorithm. The PFFLN is employed for approximating uncertainties in flight. Its weights are online tuned based on Lyapunov stability theorem for the first time. The learning process does not need any offline training phase. Additionally, a robust controller with an adaptive gain is designed to offset the approximation error. Finally, simulation results show a satisfactory performance for the NHV attitude tracking, and also illustrate the controller’s robustness.

Fault detection observer design for networked control system with long time-delays and data packet dropout
Xuan Li, Xiaobei Wu, Zhiliang Xu, and Cheng Huang
2010, 21(5):  877-882.  doi:10.3969/j.issn.1004-4132.2010.05.023
Abstract ( )   PDF (547KB) ( )  
Related Articles | Metrics

Focusing on the networked control system with long time-delays and data packet dropout, the problem of observerbased fault detection of the system is studied. According to conditions of data arrival of the controller, the state observers of the system are designed to detect faults when they occur in the system. When the system is normal, the observers system is modeled as an uncertain switched system. Based on the model, stability condition of the whole system is given. When conditions are satisfied, the system is asymptotically stable. When a fault occurs, the observers residual can change rapidly to detect the fault. A numerical example shows the effectiveness of the proposed method.

Missile robust gain scheduling autopilot design using full block multipliers
Jianqiao Yu, Guanchen Luo, and Wentao Yin
2010, 21(5):  883-891.  doi:10.3969/j.issn.1004-4132.2010.05.024
Abstract ( )   PDF (1184KB) ( )  
Related Articles | Metrics

Reduction of conservatism is one of the key and difficult problems in missile robust gain scheduling autopilot design based on multipliers. This article presents a scheme of adopting linear parameter-varying (LPV) control approach with full block multipliers to design a missile robust gain scheduling autopilot in order to eliminate conservatism. A model matching design structure with a high demand on matching precision is constructed based on the missile linear fractional transformation (LFT) model. By applying full block S-procedure and elimination lemma, a convex feasibility problem with an infinite number of constraints is formulated to satisfy robust quadratic performance specifications. Then a grid method is adopted to transform the infinite-dimensional convex feasibility problem into a solvable finite-dimensional convex feasibility problem, based on which a gain scheduling controller with linear fractional dependence on the flight Mach number and altitude is derived. Static and dynamic simulation results show the effectiveness and feasibility of the proposed scheme.

Distributed rate allocation for elastic flows in concurrent multipath transfer
Shiyong Li, Yajuan Qin, and Hongke Zhang
2010, 21(5):  892-899.  doi:10.3969/j.issn.1004-4132.2010.05.025
Abstract ( )   PDF (1051KB) ( )  
Related Articles | Metrics

Concurrent multipath transfer (CMT) using stream control transmission protocol (SCTP) multihoming has become an appealing option to increase the throughput and improve the performance of increasingly bandwidth-hungry applications. To investigate the rate allocation for applications in CMT, this paper analyzes the capacities of paths shared by competing sources, then proposes the rate allocation model for elastic flows based on the framework of network utility maximization (NUM). In order to obtain the global optimum of the model, a distributed algorithm is presented which depends only on local available information. Simulation results confirm that the proposed algorithm can achieve the global optimum within reasonable convergence times.

Robust estimation algorithm for multiple-structural data
Zhiling Wang and Zonghai Chen
2010, 21(5):  900-906.  doi:10.3969/j.issn.1004-4132.2010.05.026
Abstract ( )   PDF (1051KB) ( )  
Related Articles | Metrics

This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable (EIV) model. The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data. Under the structural density assumption, the C-step technique borrowed from the Rousseeuw’s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization. To eliminate the model ambiguities of the multiple-structural data, statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation. Experiments show that the efficiency and robustness of the proposed algorithm. This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable (EIV) model. The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data. Under the structural density assumption, the C-step technique borrowed from the Rousseeuw’s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization. To eliminate the model ambiguities of the multiple-structural data, statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation. Experiments show that the efficiency and robustness of the proposed algorithm.

Head pose estimation method based on pose manifold and tensor decomposition
Wei Wei, Yanning Zhang, and Chunna Tian
2010, 21(5):  907-913.  doi:10.3969/j.issn.1004-4132.2010.05.027
Abstract ( )   PDF (354KB) ( )  
Related Articles | Metrics

Pose manifold and tensor decomposition are used to represent the nonlinear changes of multi-view faces for pose estimation, which cannot be well handled by principal component analysis or multilinear analysis methods. A pose manifold generation method is introduced to describe the nonlinearity in pose subspace. And a nonlinear kernel based method is used to build a smooth mapping from the low dimensional pose subspace to the high dimensional face image space. Then the tensor decomposition is applied to the nonlinear mapping coefficients to build an accurate multi-pose face model for pose estimation. More importantly, this paper gives a proper distance measurement on the pose manifold space for the nonlinear mapping and pose estimation. Experiments on the identity unseen face images show that the proposed method increases pose estimation rates by 13.8% and 10.9% against principal component analysis and multilinear analysis based methods respectively. Thus, the proposed method can be used to estimate a wide range of head poses.  

Multilayer ANN indoor location system with area division in WLAN environment
Mu Zhou, Yubin Xu, and Li Tang
2010, 21(5):  914-926.  doi:10.3969/j.issn.1004-4132.2010.05.028
Abstract ( )   PDF (6356KB) ( )  
Related Articles | Metrics

An indoor location system based on multilayer artificial neural network (ANN) with area division is proposed. The characteristics of recorded signal strength (RSS), or signal to noise ratio (SNR) from each available access points (APs), are utilized to establish the radio map in the off-line phase. And in the on-line phase, the two or three dimensional coordinates of mobile terminals (MTs) are estimated according to the similarity between the new recorded RSS or SNR and fingerprints pre-stored in radio map. Although the feed-forward ANN with three layers is sufficient to describe any nonlinear mapping relationship between inputs and outputs with finite discontinuous points, the efficient inputs for better training performances are difficult to be determined because of complex and dynamic indoor environment. Then, the discussion of distance relativity for different signal characteristics and optimal strategies for multi-mode phenomenon avoidance is presented. And also, the feasibility and effectiveness of this method are verified based on the experimental comparison with normal ANN without area division, K-nearest neighbor (KNN) and probability methods in typical office environment.