Quan Pan

Northwestern Polytechnical University, Xi’an, Liaoning, China

Are you Quan Pan?

Claim your profile

Publications (255)187.39 Total impact

  • Yong Liu · Yan Liang · Zhun‐Ga Liu · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: An Interacting Multiple Model (IMM) filter suppresses outliers by integrating model-conditioned estimation and model mode recognition through adaptive mode switches based on a Bayesian probability update. However, it may face significant outlier-caused peak errors, which are undesirable or even impermissible in many applications, for example, target tracking. This paper considers that, from the view of the Dempster-Shafer Theory, the IMM filter actually fuses the Bayesian beliefs of the model mode with Dempster's Rule of Combination, which can deal with uncertainties instead of the evidence conflicts that may exist as outliers appear. Therefore, we propose the adaptive robust multiple model (RMM) filter through introducing practical expert knowledge in the belief function framework, which runs online to reduce outlier-caused peak errors. In RMM, the Likelihood Temporal Ratio (LTR) is incorporated to provide extra information on the tendency of mode switching. Moreover, expert rules are introduced to construct adaptive belief functions using discount techniques and to select the combination rule that better deals with evidence conflicts when outliers appear. The simulations in target tracking scenarios show that the proposed RMM filter significantly reduces outlier-caused peak errors and hence obtains a lower rate of track loss in the cluttered environment.
    Asian Journal of Control 11/2015; DOI:10.1002/asjc.1042 · 1.41 Impact Factor
  • Kuang Zhou · Arnaud Martin · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: Communities are of great importance for understanding graph structures in social networks. Some existing community detection algorithms use a single prototype to represent each group. In real applications, this may not adequately model the different types of communities and hence limits the clustering performance on social networks. To address this problem, a Similarity-based Multi-Prototype (SMP) community detection approach is proposed in this paper. In SMP, vertices in each community carry various weights to describe their degree of representativeness. This mechanism enables each community to be represented by more than one node. The centrality of nodes is used to calculate prototype weights, while similarity is utilized to guide us to partitioning the graph. Experimental results on computer generated and real-world networks clearly show that SMP performs well for detecting communities. Moreover, the method could provide richer information for the inner structure of the detected communities with the help of prototype weights compared with the existing community detection models.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In real clustering applications, proximity data, in which only pairwise similarities or dissimilarities are known, is more general than object data, in which each pattern is described explicitly by a list of attributes. Medoid-based clustering algorithms, which assume the prototypes of classes are objects, are of great value for partitioning relational data sets. In this paper a new prototype-based clustering method, named Evidential C-Medoids (ECMdd), which is an extension of Fuzzy C-Medoids (FCMdd) on the theoretical framework of belief functions is proposed. In ECMdd, medoids are utilized as the prototypes to represent the detected classes, including specific classes and imprecise classes. Specific classes are for the data which are distinctly far from the prototypes of other classes, while imprecise classes accept the objects that may be close to the prototypes of more than one class. This soft decision mechanism could make the clustering results more cautious and reduce the misclassification rates. Experiments in synthetic and real data sets are used to illustrate the performance of ECMdd. The results show that ECMdd could capture well the uncertainty in the internal data structure. Moreover, it is more robust to the initializations compared with FCMdd.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Among the computational intelligence techniques employed to solve classification problems, the fuzzy rule-based classification system (FRBCS) is a popular tool capable of building a linguistic model interpretable to users. However, it may face lack of accuracy in some complex applications, by the fact that the inflexibility of the concept of the linguistic variable imposes hard restrictions on the fuzzy rule structure. In this paper, we extend the fuzzy rule in FRBCS with a belief rule structure and develop a belief rule-based classification system (BRBCS) to address imprecise or incomplete information in complex classification problems. The two components of the proposed BRBCS, i.e., the belief rule base (BRB) and the belief reasoning method (BRM), are designed specifically by taking into account the pattern noise that existes in many real-world data sets. Four experiments based on benchmark data sets are carried out to evaluate the classification accuracy, robustness, interpretability and time complexity of the proposed method.
    Information Sciences 07/2015; 309. DOI:10.1016/j.ins.2015.03.005 · 3.89 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The influence of the missing values in the classification of incomplete pattern mainly depends on the context. In this paper, we present a fast classification method for incomplete pattern based on the fusion of belief functions where the missing values are selectively (adaptively) estimated. At first, it is assumed that the missing information is not crucial for the classification, and the object (incomplete pattern) is classified based only on the available attribute values. However, if the object cannot be clearly classified, it implies that the missing values play an important role to obtain an accurate classification. In this case, the missing values will be imputed based on the K-nearest neighbor (K-NN) and self-organizing map (SOM) techniques, and the edited pattern with the imputation is then classified. The (original or edited) pattern is respectively classified according to each training class, and the classification results represented by basic belief assignments (BBA's) are fused with proper combination rules for making the credal classification. The object is allowed to belong with different masses of belief to the specific classes and meta-classes (i.e. disjunctions of several single classes). This credal classification captures well the uncertainty and imprecision of classification, and reduces effectively the rate of misclassifications thanks to the introduction of meta-classes. The effectiveness of the proposed method with respect to other classical methods is demonstrated based on several experiments using artificial and real data sets.
    Fusion 2015, Washington DC; 07/2015
  • Zhun-ga Liu · Yong Liu · Jean Dezert · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: It can be quite difficult to correctly and precisely classify the incomplete data with missing values, since the missing information usually causes ambiguities (uncertainty) in the classification result. Belief function theory can well model such uncertain and imprecise information, and a new belief-based method for credal classification of incomplete data (CCI) is proposed using the K nearest neighbors (KNNs) strategy. In CCI, the KNNs of object (incomplete data) are respectively used to estimate the missing values, and one can obtain K versions of edited pattern with estimated values from the KNNs. The K edited patterns are classified by any classical method to get K pieces of classification results with different discounting (weighting) factors depending on the distances between the object and its KNNs, and global fusion of the K classification results represented by the basic belief assignments (bba’s) is used for credal classification of the object. The conflicting beliefs produced in the fusion process can well capture the imprecision degree of classification, and it will be transferred to the selected meta-class defined by the disjunction of several classes (i.e. the set of several classes) according to the current context. Thus, the incomplete data that is hard to correctly classify because of the missing values will be reasonably committed to proper meta-class, which is able to characterize the imprecision of classification and reduce the errors as well. Three experiments are given to illustrate the potential and interest of CCI approach.
  • Kuang Zhou · Arnaud Martin · Quan Pan
  • Lianmeng Jiao · Quan Pan · Xiaoxue Feng
    [Show abstract] [Hide abstract]
    ABSTRACT: The performance of nearest-neighbor (NN) classifiers is known to be very sensitive to the distance metric used in classifying a query pattern, especially in scarce-prototype cases. In this paper, a class-conditional weighted (CCW) distance metric related to both the class labels of the prototypes and the query patterns is proposed. Compared with the existing distance metrics, the proposed metric provides more flexibility to design the feature weights so that the local specifics in feature space can be well characterized. Based on the proposed CCW distance metric, a multi-hypothesis nearest-neighbor (MHNN) classifier is developed. The scheme of the proposed MHNN classifier is to classify the query pattern under multiple hypotheses in which the nearest-neighbor sub-classifiers can be implemented based on the CCW distance metric. Then the classification results of multiple sub-classifiers are combined to get the final result. Under this general scheme, a specific realization of the MHNN classifier is developed within the framework of Dempster-Shafer theory due to its good capability of representing and combining uncertain information. Two experiments based on synthetic and real data sets were carried out to show the effectiveness of the proposed technique.
    Neurocomputing 03/2015; 151:1468-1476. DOI:10.1016/j.neucom.2014.10.039 · 2.01 Impact Factor
  • Source
    Kuang Zhou · Arnaud Martin · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: Community detection is of great importance for understand-ing graph structure in social networks. The communities in real-world networks are often overlapped, i.e. some nodes may be a member of multiple clusters. How to uncover the overlapping communities/clusters in a complex network is a general problem in data mining of network data sets. In this paper, a novel algorithm to identify overlapping communi-ties in complex networks by a combination of an evidential modularity function, a spectral mapping method and evidential c-means clustering is devised. Experimental results indicate that this detection approach can take advantage of the theory of belief functions, and preforms good both at detecting community structure and determining the appropri-ate number of clusters. Moreover, the credal partition obtained by the proposed method could give us a deeper insight into the graph structure.
    Communications in Computer and Information Science 01/2015; 442. DOI:10.1007/978-3-319-08795-5_57
  • Source
    Kuang Zhou · Arnaud Martin · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: Evidential-EM (E2M) algorithm is an effective approach for computing maximum likelihood estimations under finite mixture models, especially when there is uncertain information about data. In this paper we present an extension of the E2M method in a particular case of incom-plete data, where the loss of information is due to both mixture models and censored observations. The prior uncertain information is expressed by belief functions, while the pseudo-likelihood function is derived based on imprecise observations and prior knowledge. Then E2M method is evoked to maximize the generalized likelihood function to obtain the optimal estimation of parameters. Numerical examples show that the proposed method could effectively integrate the uncertain prior infor-mation with the current imprecise knowledge conveyed by the observed data.
  • Xiaoxu Wang · Quan Pan · Yan Liang · Hanzhou Li
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper focuses on designing Gaussian sum approximation filter for both accurately and rapidly estimating the state of a class of nonlinear dynamic time-delay systems. Firstly, a novel nonaugmented Gaussian filter (GF) is derived, whose superiority in computation efficiency is theoretically analyzed as compared to the standard augmented GF. Secondly, a nonaugmented Gaussian sum filter (GSF) is proposed to accurately capture the state estimates by a weight sum of the above-proposed GF. In GSF, each GF component is independent from the others and can be performed in a parallel manner so that GSF is conducive to high-performance computing across many compute nodes. Finally, the performance of the proposed GSF is demonstrated by a vehicle suspension system with time delay, where the GSF achieves higher accuracy than the single GF and is computationally much more efficient than the particle filter with the almost same accuracy.
    Nonlinear Dynamics 01/2015; DOI:10.1007/s11071-015-2171-5 · 2.42 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The recent credal partition approach allows the objects to belong to not only the singleton clusters but also the sets of clusters (i.e. meta-clusters) with different masses of belief. A new credal c-means (CCM) clustering method working with credal partition has been proposed in this work to effectively deal with the uncertain and imprecise data. In the clustering problem, one object simultaneously close to several clusters can be difficult to correctly classify, since these close clusters appear not very distinguishable for this object. In such case, the object will be cautiously committed by CCM to a meta-cluster (i.e. the disjunction of these close clusters), which can be considered as a transition cluster among these different close clusters. It can well characterize the imprecision of the class of the object and can also reduce the misclassification errors thanks to the use of meta-cluster. CCM is robust to the noisy data because of the outlier cluster. The clustering centers and the mass of belief on each cluster for any object are obtained by the optimization of a proper objective function in CCM. The effectiveness of CCM has been demonstrated by three experiments using synthetic and real data sets with respect to fuzzy c-means (FCM) and evidential c-means (ECM) clustering methods.
    Knowledge-Based Systems 11/2014; 74(1). DOI:10.1016/j.knosys.2014.11.013 · 3.06 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Median clustering is of great value for partitioning relational data. In this paper, a new prototype-based clustering method, called Median Evidential C-Means (MECM), which is an extension of median c-means and median fuzzy c-means on the theoretical framework of belief functions is proposed. The median variant relaxes the restriction of a metric space embedding for the objects but constrains the prototypes to be in the original data set. Due to these properties, MECM could be applied to graph clustering problems. A community detection scheme for social networks based on MECM is investigated and the obtained credal partitions of graphs, which are more refined than crisp and fuzzy ones, enable us to have a better understanding of the graph structures. An initial prototype-selection scheme based on evidential semi-centrality is presented to avoid local premature convergence and an evidential modularity function is defined to choose the optimal number of communities. Finally, experiments in synthetic and real data sets illustrate the performance of MECM and show its difference to other methods.
    Knowledge-Based Systems 11/2014; 74. DOI:10.1016/j.knosys.2014.11.010 · 3.06 Impact Factor
  • Xiaoxu Wang · Yan Liang · Quan Pan · Zengfu Wang
    [Show abstract] [Hide abstract]
    ABSTRACT: A recent comment (Chang, 2014) theoretically demonstrated that for the linear system with the correlated process and measurement noises, the GASF framework proposed in the paper Wang et al. (2012) was equivalent to the conventional de-coupling filtering framework. In this note, we would further show that such equivalence between the two frameworks can be justified in a more general way, even for the nonlinear system.
    Automatica 10/2014; 50(12). DOI:10.1016/j.automatica.2014.10.040 · 3.13 Impact Factor
  • Lin Zhou · Yan Liang · Jie Zhou · Feng Yang · Quan Pan
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, a generalised systematic bias (SB) is presented, which is represented via a dynamic model driven by structured unknown inputs (UI). The online SB estimation is implemented in two steps. In the first step, the state-free SB measurement and the UI-free SB dynamic model are derived in the case that UI-free condition holds. In the second step, the linear minimum mean squared filter is obtained via orthogonal principle, and the sufficient condition of filtering stability is presented. A simulation about target tracking is given to verify the proposed method.
    IET Radar Sonar ? Navigation 10/2014; 8(8):977-986. DOI:10.1049/iet-rsn.2013.0311 · 1.03 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A dynamic attitude measurement system (DAMS) is developed based on a laser inertial navigation system (LINS). Three factors of the dynamic attitude measurement error using LINS are analyzed: dynamic error, time synchronization and phase lag. An optimal coning errors compensation algorithm is used to reduce coning errors, and two-axis wobbling verification experiments are presented in the paper. The tests indicate that the attitude accuracy is improved 2-fold by the algorithm. In order to decrease coning errors further, the attitude updating frequency is improved from 200 Hz to 2000 Hz. At the same time, a novel finite impulse response (FIR) filter with three notches is designed to filter the dither frequency of the ring laser gyro (RLG). The comparison tests suggest that the new filter is five times more effective than the old one. The paper indicates that phase-frequency characteristics of FIR filter and first-order holder of navigation computer constitute the main sources of phase lag in LINS. A formula to calculate the LINS attitude phase lag is introduced in the paper. The expressions of dynamic attitude errors induced by phase lag are derived. The paper proposes a novel synchronization mechanism that is able to simultaneously solve the problems of dynamic test synchronization and phase compensation. A single-axis turntable and a laser interferometer are applied to verify the synchronization mechanism. The experiments results show that the theoretically calculated values of phase lag and attitude error induced by phase lag can both match perfectly with testing data. The block diagram of DAMS and physical photos are presented in the paper. The final experiments demonstrate that the real-time attitude measurement accuracy of DAMS can reach up to 20″ (1σ) and the synchronization error is less than 0.2 ms on the condition of three axes wobbling for 10 min.
    Sensors 09/2014; 14(9):16082-16108. DOI:10.3390/s140916082 · 2.05 Impact Factor
  • Feng Yang · Yuemei Qin · Yan Liang · Quan Pan · Yanbo Yang
    IET Control Theory and Applications 08/2014; 8(12):1112-1126. DOI:10.1049/iet-cta.2013.0936 · 1.84 Impact Factor
  • Yanbo Yang · Yuemei Qin · Feng Yang · Quan Pan · Yan Liang
    [Show abstract] [Hide abstract]
    ABSTRACT: This study presents the state estimation problem of discrete-time Markovian jump linear systems with randomly delayed measurements. Here, the delay is modelled as the combination of different number of binary stochastic variables according to the different possible delay steps. In the actually delayed measurement equation, multiple adjacent step measurement noises are correlated. Owing to the stochastic property from the measurement delay, the estimation model is rewritten as a discrete-time system with stochastic parameters and augmented state reconstructed from all modes with their mode uncertainties. For this system, a novel linear minimum-mean-square error (LMMSE, renamed as LMRDE) estimator for the augmented state is derived in a recursive structure according to the orthogonality principle under a generalised framework. Since the correlation among multiple adjacent step noises in the measurement equation, the measurement noises and related second moment matrices of corresponding previous instants in each current step are also needed to be estimated or calculated. A numerical example with possibly delayed measurements is simulated to testify the proposed method.
    IET Signal Processing 08/2014; 8(6):658-667. DOI:10.1049/iet-spr.2013.0431 · 0.69 Impact Factor
  • Xiaoxu Wang · Yan Liang · Quan Pan · He Huang
    [Show abstract] [Hide abstract]
    ABSTRACT: The Gaussian mixture approximation to the probability density function of the state is more appropriate than the single Gaussian approximation. A Gaussian mixture filter (GMF) is proposed for a class of non-linear discrete-time stochastic systems with the multi-state delayed case. First, a novel non-augmented filtering framework of the constituent Gaussian filter (GF) in GMF is derived, which recursively operates by analytical computation and non-linear Gaussian integrals. The implementation of such GF is thus transformed to the computation of such non-linear integrals in the proposed framework, which is solved by applying different numerical technologies for developing various variations of the non-augmented GF, for example, GF-cubature Kalman filter (CKF) based on the cubature rule. Secondly, a non-augmented GMF is discussed by a weight sum of the above proposed GF, where each GF component is independent from the others and can be performed in a parallel manner, and its corresponding weigh is updated by using the measurements according to Bayesian formula. Naturally, a variation or implementation of such GMF based on the cubature rule is the GMF-CKF. Finally, the performance of the new filters is demonstrated by a numerical example and a vehicle suspension estimation problem.
    IET Control Theory and Applications 07/2014; 8(11):996-1008. DOI:10.1049/iet-cta.2013.0875 · 1.84 Impact Factor
  • Source
    Liu Z. · Pan Q. · Mercier G. · Dezert J.
    [Show abstract] [Hide abstract]
    ABSTRACT: The missing data in incomplete pattern can have different estimations, and the classification result of pattern with different estimations may be quite distinct. Such uncertainty (ambiguity) of classification is mainly caused by the loss of information in missing data. A new prototype-based credal classification (PCC) method is proposed to classify incomplete patterns using belief functions. The class prototypes obtained by the training data are respectively used to estimate the missing values. Typically, in a c-class problem, one has to deal with c prototypes which yields c estimations. The different edited patterns based on each possible estimation are then classified by a standard classifier and one can get c classification results for an incomplete pattern. Because all these classification results are potentially admissible, they are fused altogether to obtain the credal classification of the incomplete pattern. A new credal combination method is introduced for solving the classification problem, and it is able to characterize the inherent uncertainty due to the possible conflicting results delivered by the different estimations of missing data. The incomplete patterns that are hard to correctly classify will be reasonably committed to some proper meta-classes by PCC method in order to reduce the misclassification rate. The use and potential of PCC method is illustrated through several experiments with artificial and real data sets.
    Fusion 2014 Int Conf on Information Fusion, Salamanca, Spain; 07/2014

Publication Stats

2k Citations
187.39 Total Impact Points

Institutions

  • 1994–2015
    • Northwestern Polytechnical University
      • School of Automation
      Xi’an, Liaoning, China
  • 2013
    • University of Melbourne
      • Department of Electrical and Electronic Engineering
      Melbourne, Victoria, Australia
  • 2010
    • University of Alberta
      • Department of Electrical and Computer Engineering
      Edmonton, Alberta, Canada
  • 2004–2010
    • Northwestern Polytechnic University
      China, Maine, United States
  • 2008
    • Shanghai Jiao Tong University
      • State Key Laboratory of Medical Genomics
      Shanghai, Shanghai Shi, China
    • Ruijin Hospital North
      Shanghai, Shanghai Shi, China
  • 2007
    • Nanjing Institute of Technology
      Nan-ching-hsü, Jiangxi Sheng, China
    • Harvard Medical School
      • Department of Biological Chemistry and Molecular Pharmacology
      Boston, Massachusetts, United States
  • 2006
    • Beijing Normal University
      • State Key Laboratory of Remote Sensing Sciences
      Peping, Beijing, China
  • 2005
    • Robert-Bosch Krankenhaus
      Stuttgart, Baden-Württemberg, Germany
  • 2001
    • University of Illinois, Urbana-Champaign
      • Department of Electrical and Computer Engineering
      Urbana, Illinois, United States