Showing papers in "International Journal of Applied Mathematics and Computer Science in 2014"

A support vector machine with the tabu search algorithm for freeway incident detection

Abstract: Automated Incident Detection AID is an important part of Advanced Trafźc Management and Information Systems ATMISs. An automated incident detection system can effectively provide information on an incident, which can help initiate the required measure to reduce the inźuence of the incident. To accurately detect incidents in expressways, a Support Vector Machine SVM is used in this paper. Since the selection of optimal parameters for the SVM can improve prediction accuracy, the tabu search algorithm is employed to optimize the SVM parameters. The proposed model is evaluated with data for two freeways in China. The results show that the tabu search algorithm can effectively provide better parameter values for the SVM, and SVM models outperform Artiźcial Neural Networks ANNs in freeway incident detection.

Disturbance modeling and state estimation for offset-free predictive control with state-space process models

Abstract: Disturbance modeling and design of state estimators for offset-free Model Predictive Control MPC with linear state-space process models is considered in the paper for deterministic constant-type external and internal disturbances modeling errors. The application and importance of constant state disturbance prediction in the state-space MPC controller design is presented. In the case with a measured state, this leads to the control structure without disturbance state observers. In the case with an unmeasured state, a new, simpler MPC controller-observer structure is proposed, with observation of a pure process state only. The structure is not only simpler, but also with less restrictive applicability conditions than the conventional approach with extended process-and-disturbances state estimation. Theoretical analysis of the proposed structure is provided. The design approach is also applied to the case with an augmented state-space model in complete velocity form. The results are illustrated on a 2×2 example process problem.

On truncations for weakly ergodic inhomogeneous birth and death processes

Abstract: We investigate a class of exponentially weakly ergodic inhomogeneous birth and death processes. We consider special transformations of the reduced intensity matrix of the process and obtain uniform in time error bounds of truncations. Our approach also guarantees that we can find limiting characteristics approximately with an arbitrarily fixed error. As an example, we obtain the respective bounds of the truncation error for an Mt/Mt/S queue for any number of servers S. Arbitrary intensity functions instead of periodic ones can be considered in the same manner.

A finite-buffer queue with a single vacation policy

Abstract: In this paper, application of an evolutionary strategy to positioning a GI/M/1/N-type finite-buffer queueing system with exhaustive service and a single vacation policy is presented. The examined object is modeled by a conditional joint transform of the first busy period, the first idle time and the number of packets completely served during the first busy period. A mathematical model is defined recursively by means of input distributions. In the paper, an analytical study and numerical experiments are presented. A cost optimization problem is solved using an evolutionary strategy for a class of queueing systems described by exponential and Erlang distributions.

Approximation of phenol concentration using novel hybrid computational intelligence methods

Abstract: This paper presents two innovative evolutionary-neural systems based on feed-forward and recurrent neural networks used for quantitative analysis. These systems have been applied for approximation of phenol concentration. Their performance was compared against the conventional methods of artificial intelligence (artificial neural networks, fuzzy logic and genetic algorithms). The proposed systems are a combination of data preprocessing methods, genetic algorithms and the Levenberg–Marquardt (LM) algorithm used for learning feed forward and recurrent neural networks. The initial weights and biases of neural networks chosen by the use of a genetic algorithm are then tuned with an LM algorithm. The evaluation is made on the basis of accuracy and complexity criteria. The main advantage of proposed systems is the elimination of random selection of the network weights and biases, resulting in increased efficiency of the systems.

Nuclei segmentation for computer-aided diagnosis of breast cancer

Abstract: Breast cancer is the most common cancer among women. The effectiveness of treatment depends on early detection of the disease. Computer-aided diagnosis plays an increasingly important role in this field. Particularly, digital pathology has recently become of interest to a growing number of scientists. This work reports on advances in computer-aided breast cancer diagnosis based on the analysis of cytological images of fine needle biopsies. The task at hand is to classify those as either benign or malignant. We propose a robust segmentation procedure giving satisfactory nuclei separation even when they are densely clustered in the image. Firstly, we determine centers of the nuclei using conditional erosion. The erosion is performed on a binary mask obtained with the use of adaptive thresholding in grayscale and clustering in a color space. Then, we use the multi-label fast marching algorithm initialized with the centers to obtain the final segmentation. A set of 84 features extracted from the nuclei is used in the classification by three different classifiers. The approach was tested on 450 microscopic images of fine needle biopsies obtained from patients of the Regional Hospital in Zielona Gora, Poland. The classification accuracy presented in this paper reaches 100%, which shows that a medical decision support system based on our method would provide accurate diagnostic information.

A new lightweight method for security risk assessment based on fuzzy cognitive maps

Abstract: For contemporary software systems, security is considered to be a key quality factor and the analysis of IT security risk becomes an indispensable stage during software deployment. However, performing risk assessment according to methodologies and standards issued for the public sector or large institutions can be too costly and time consuming. Current business practice tends to circumvent risk assessment by defining sets of standard safeguards and applying them to all developed systems. This leads to a substantial gap: threats are not re-evaluated for particular systems and the selection of security functions is not based on risk models. This paper discusses a new lightweight risk assessment method aimed at filling this gap. In this proposal, Fuzzy Cognitive Maps (FCMs) are used to capture dependencies between assets, and FCM-based reasoning is performed to calculate risks. An application of the method is studied using an example of an e-health system providing remote telemonitoring, data storage and teleconsultation services. Lessons learned indicate that the proposed method is an efficient and low-cost approach, giving instantaneous feedback and enabling reasoning on the effectiveness of the security system.

Forecasting return products in an integrated forward/reverse supply chain utilizing an ANFIS

Abstract: Interests in Closed-Loop Supply Chain CLSC issues are growing day by day within the academia, companies, and customers. Many papers discuss profitability or cost reduction impacts of remanufacturing, but a very important point is almost missing. Indeed, there is no guarantee about the amounts of return products even if we know a lot about demands of first products. This uncertainty is due to reasons such as companies' capabilities in collecting End-of-Life EOL products, customers' interests in returning and current incentives, and other independent collectors. The aim of this paper is to deal with the important gap of the uncertainties of return products. Therefore, we discuss the forecasting method of return products which have their own open-loop supply chain. We develop an integrated two-phase methodology to cope with the closed-loop supply chain design and planning problem. In the first phase, an Adaptive Network Based Fuzzy Inference System ANFIS is presented to handle the uncertainties of the amounts of return product and to determine the forecasted return rates. In the second phase, and based on the results of the first one, the proposed multi-echelon, multi-product, multi-period, closed-loop supply chain network is optimized. The second-phase optimization is undertaken based on using general exact solvers in order to achieve the global optimum. Finally, the performance of the proposed forecasting method is evaluated in 25 periods using a numerical example, which contains a pattern in the returning of products. The results reveal acceptable performance of the proposed two-phase optimization method. Based on them, such forecasting approaches can be applied to real-case CLSC problems in order to achieve more reliable design and planning of the network

Minimum energy control of fractional descriptor positive discrete-time linear systems

Abstract: Necessary and sufficient conditions for the positivity and reachability of fractional descriptor positive discrete-time linear systems are established. The minimum energy control problem for descriptor positive systems is formulated and solved. Sufficient conditions for the existence of a solution to the minimum energy control problem are given. A procedure for computation of optimal input sequences and a minimal value of the performance index is proposed and illustrated by a numerical example.

A Differential Evolution Approach to Dimensionality Reduction for Classification Needs

Abstract: The feature selection problem often occurs in pattern recognition, and more specific, classification. Although these patterns could contain a large number of features, some of them could prove to be irrelevant, redundant or even detrimental to classification accuracy. Thus, it is important to remove these kinds of features which in turn leads to problem dimensionality reduction and could eventually improve the classification accuracy. In this paper an approach to dimensionality reduction based on differential evolution which represents a wrapper and explores the solution space is presented. The solutions, subsets of the whole feature set, are evaluated using the k-nearest neighbour algorithm. High quality solutions found during execution of the differential evolution fill the archive. A final solution is obtained by conducting k-fold cross validation on the archive solutions and selecting the best. Experimental analysis was conducted on several standard test sets. Classification accuracy of the k-nearest neighbour algorithm using the full feature set and the accuracy of the same algorithm using only the subset provided by the proposed approach and some other optimization algorithms which were used as wrappers are compared. The analysis has shown that the proposed approach successfully determines good feature subsets which may increase classification accuracy.

Analysis of an M|G|1|R queue with batch arrivals and two hysteretic overload control policies

Abstract: Hysteretic control of arrivals is one of the most easy-to-implement and effective solutions of overload problems occurring in SIP-servers. A mathematical model of an SIP server based on the queueing system M[X]|G|1L,H|H,R with batch arrivals and two hysteretic loops is being analyzed. This paper proposes two analytical methods for studying performance characteristics related to the number of customers in the system. Two control policies defined by instants when it is decided to change the system's mode are considered. The expression for an important performance characteristic of each policy the mean time between changes in the system mode is presented. Numerical examples that allow comparison of the efficiency of both policies are given

From the Slit-Island Method to the Ising model: Analysis of irregular grayscale objects

Abstract: The Slit Island Method (SIM) is a technique for the estimation of the fractal dimension of an object by determining the area– perimeter relations for successive slits. The SIM could be applied for image analysis of irregular grayscale objects and their classification using the fractal dimension. It is known that this technique is not functional in some cases. It is emphasized in this paper that for specific objects a negative or an infinite fractal dimension could be obtained. The transformation of the input image data from unipolar to bipolar gives a possibility of reformulated image analysis using the Ising model context. The polynomial approximation of the obtained area-perimeter curve allows object classification. The proposed technique is applied to the images of cervical cell nuclei (Papanicolaou smears) for the preclassification of the correct and atypical cells.

Minimum energy control of fractional positive continuous-time linear systems with bounded inputs

Abstract: A minimum energy control problem for fractional positive continuous-time linear systems with bounded inputs is formulated and solved. Sufźcient conditions for the existence of a solution to the problem are established. A procedure for solving the problem is proposed and illustrated with a numerical example.

A hybrid algorithm for solving inverse problems in elasticity

Abstract: The paper offers a new approach to handling difficult parametric inverse problems in elasticity and thermo-elasticity, formulated as global optimization ones. The proposed strategy is composed of two phases. In the first, global phase, the stochastic hp-HGS algorithm recognizes the basins of attraction of various objective minima. In the second phase, the local objective minimizers are closer approached by steepest descent processes executed singly in each basin of attraction. The proposed complex strategy is especially dedicated to ill-posed problems with multimodal objective functionals. The strategy offers comparatively low computational and memory costs resulting from a double-adaptive technique in both forward and inverse problem domains. We provide a result on the Lipschitz continuity of the objective functional composed of the elastic energy and the boundary displacement misfits with respect to the unknown constitutive parameters. It allows common scaling of the accuracy of solving forward and inverse problems, which is the core of the introduced double-adaptive technique. The capability of the proposed method of finding multiple solutions is illustrated by a computational example which consists in restoring all feasible Young modulus distributions minimizing an objective functional in a 3D domain of a photo polymer template obtained during step and flash imprint lithography.

Automatic speech signal segmentation based on the innovation adaptive filter

Abstract: Speech segmentation is an essential stage in designing automatic speech recognition systems and one can źnd several algorithms proposed in the literature. It is a difźcult problem, as speech is immensely variable. The aim of the authors' studies was to design an algorithm that could be employed at the stage of automatic speech recognition. This would make it possible to avoid some problems related to speech signal parametrization. Posing the problem in such a way requires the algorithm to be capable of working in real time. The only such algorithm was proposed by Tyagi et al., 2006, and it is a modiźed version of Brandt's algorithm. The article presents a new algorithm for unsupervised automatic speech signal segmentation. It performs segmentation without access to information about the phonetic content of the utterances, relying exclusively on second-order statistics of a speech signal. The starting point for the proposed method is time-varying Schur coefźcients of an innovation adaptive źlter. The Schur algorithm is known to be fast, precise, stable and capable of rapidly tracking changes in second order signal statistics. A transfer from one phoneme to another in the speech signal always indicates a change in signal statistics caused by vocal track changes. In order to allow for the properties of human hearing, detection of inter-phoneme boundaries is performed based on statistics deźned on the mel spectrum determined from the reźection coefźcients. The paper presents the structure of the algorithm, deźnes its properties, lists parameter values, describes detection efźciency results, and compares them with those for another algorithm. The obtained segmentation results, are satisfactory.

An algorithm for reducing the dimension and size of a sample for data exploration procedures

Abstract: The paper deals with the issue of reducing the dimension and size of a data set (random sample) for exploratory data analysis procedures. The concept of the algorithm investigated here is based on linear transformation to a space of a smaller dimension, while retaining as much as possible the same distances between particular elements. Elements of the transformation matrix are computed using the metaheuristics of parallel fast simulated annealing. Moreover, elimination of or a decrease in importance is performed on those data set elements which have undergone a significant change in location in relation to the others. The presented method can have universal application in a wide range of data exploration problems, offering flexible customization, possibility of use in a dynamic data environment, and comparable or better performance with regards to the principal component analysis. Its positive features were verified in detail for the domain’s fundamental tasks of clustering, classification and detection of atypical elements (outliers).

Further results on robust fuzzy dynamic systems with LMI D-stability constraints

Abstract: This paper examines the problem of designing a robust H∞ fuzzy controller with D-stability constraints for a class of nonlinear dynamic systems which is described by a Takagi-Sugeno TS fuzzy model. Fuzzy modelling is a multi-model approach in which simple sub-models are combined to determine the global behavior of the system. Based on a linear matrix inequality LMI approach, we develop a robust H∞ fuzzy controller that guarantees i the L2-gain of the mapping from the exogenous input noise to the regulated output to be less than some prescribed value, and ii the closed-loop poles of each local system to be within a specified stability region. Sufficient conditions for the controller are given in terms of LMIs. Finally, to show the effectiveness of the designed approach, an example is provided to illustrate the use of the proposed methodology.

A discrete-time system with service control and repairs

Abstract: This paper discusses a discrete-time queueing system with starting failures in which an arriving customer follows three different strategies. Two of them correspond to the LCFS Last Come First Served discipline, in which displacements or expulsions of customers occur. The third strategy acts as a signal, that is, it becomes a negative customer. Also examined is the possibility of failures at each service commencement epoch. We carry out a thorough study of the model, deriving analytical results for the stationary distribution. We obtain the generating functions of the number of customers in the queue and in the system. The generating functions of the busy period as well as the sojourn times of a customer at the server, in the queue and in the system, are also provided. We present the main performance measures of the model. The versatility of this model allows us to mention several special cases of interest. Finally, we prove the convergence to the continuous-time counterpart and give some numerical results that show the behavior of some performance measures with respect to the most significant parameters of the system

A 2D system approach to the design of a robust modified repetitive-control system with a dynamic output-feedback controller

Abstract: This paper is concerned with the problem of designing a robust modiźed repetitive-control system with a dynamic output feedback controller for a class of strictly proper plants. Employing the continuous lifting technique, a continuous-discrete two-dimensional 2D model is built that accurately describes the features of repetitive control. The 2D control input contains the direct sum of the effects of control and learning, which allows us to adjust control and learning preferentially. The singular-value decomposition of the output matrix and Lyapunov stability theory are used to derive an asymptotic stability condition based on a Linear Matrix Inequality LMI. Two tuning parameters in the LMI manipulate the preferential adjustment of control and learning. A numerical example illustrates the tuning procedure and demonstrates the effectiveness of the method.

A system for deduction-based formal verification of workflow-oriented software models

Abstract: The work concerns formal verification of workflow-oriented software models using the deductive approach. The formal correctness of a model's behaviour is considered. Manually building logical specifications, which are regarded as a set of temporal logic formulas, seems to be a significant obstacle for an inexperienced user when applying the deductive approach. A system, along with its architecture, for deduction-based verification of workflow-oriented models is proposed. The process inference is based on the semantic tableaux method, which has some advantages when compared with traditional deduction strategies. The algorithm for automatic generation of logical specifications is proposed. The generation procedure is based on predefined workflow patterns for BPMN, which is a standard and dominant notation for the modeling of business processes. The main idea behind the approach is to consider patterns, defined in terms of temporal logic, as a kind of logical primitives which enable the transformation of models to temporal logic formulas constituting a logical specification. Automation of the generation process is crucial for bridging the gap between the intuitiveness of deductive reasoning and the difficulty of its practical application when logical specifications are built manually. This approach has gone some way towards supporting, hopefully enhancing, our understanding of deduction-based formal verification of workflow-oriented models.

On infinite horizon active fault diagnosis for a class of non-linear non-Gaussian systems

Abstract: The paper considers the problem of active fault diagnosis for discrete-time stochastic systems over an infinite time horizon. It is assumed that the switching between a fault-free and finitely many faulty conditions can be modelled by a finite-state Markov chain and the continuous dynamics of the observed system can be described for the fault-free and each faulty condition by non-linear non-Gaussian models with a fully observed continuous state. The design of an optimal active fault detector that generates decisions and inputs improving the quality of detection is formulated as a dynamic optimization problem. As the optimal solution obtained by dynamic programming requires solving the Bellman functional equation, approximate techniques are employed to obtain a suboptimal active fault detector.

Imitation learning of car driving skills with decision trees and random forests

Abstract: Machine learning is an appealing and useful approach to creating vehicle control algorithms, both for simulated and real vehicles One common learning scenario that is often possible to apply is learning by imitation, in which the behavior of an exemplary driver provides training instances for a supervised learning algorithm This article follows this approach in the domain of simulated car racing, using the TORCS simulator In contrast to most prior work on imitation learning, a symbolic decision tree knowledge representation is adopted, which combines potentially high accuracy with human readability, an advantage that can be important in many applications Decision trees are demonstrated to be capable of representing high quality control models, reaching the performance level of sophisticated pre-designed algorithms This is achieved by enhancing the basic imitation learning scenario to include active retraining, automatically triggered on control failures It is also demonstrated how better stability and generalization can be achieved by sacrificing human-readability and using decision tree model ensembles The methodology for learning control models contributed by this article can be hopefully applied to solve real-world control tasks, as well as to develop video game bots

Center-based l1–clustering method

Abstract: Abstract In this paper, we consider the l1-clustering problem for a finite data-point set which should be partitioned into k disjoint nonempty subsets. In that case, the objective function does not have to be either convex or differentiable, and generally it may have many local or global minima. Therefore, it becomes a complex global optimization problem. A method of searching for a locally optimal solution is proposed in the paper, the convergence of the corresponding iterative process is proved and the corresponding algorithm is given. The method is illustrated by and compared with some other clustering methods, especially with the l2-clustering method, which is also known in the literature as a smooth k-means method, on a few typical situations, such as the presence of outliers among the data and the clustering of incomplete data. Numerical experiments show in this case that the proposed l1-clustering algorithm is faster and gives significantly better results than the l2-clustering algorithm.

Analysis of an mmap/ph1, ph2/n/∞ queueing system operating in a random environment

Abstract: A multi-server queueing system with two types of customers and an infinite buffer operating in a random environment as a model of a contact center is investigated. The arrival flow of customers is described by a marked Markovian arrival process. Type 1 customers have a non-preemptive priority over type 2 customers and can leave the buffer due to a lack of service. The service times of different type customers have a phase-type distribution with different parameters. To facilitate the investigation of the system we use a generalized phase-type service time distribution. The criterion of ergodicity for a multi-dimensional Markov chain describing the behavior of the system and the algorithm for computation of its steady-state distribution are outlined. Some key performance measures are calculated. The Laplace–Stieltjes transforms of the sojourn and waiting time distributions of priority and non-priority customers are derived. A numerical example illustrating the importance of taking into account the correlation in the arrival process is presented.

A modified van der Pol equation with delay in a description of the heart action

Abstract: In this paper, a modified van der Pol equation is considered as a description of the heart action. This model has a number of interesting properties allowing reconstruction of phenomena observed in physiological experiments as well as in Holter electrocardiographic recordings. Our aim is to study periodic solutions of the modified van der Pol equation and take into consideration the influence of feedback and delay which occur in the normal heart action mode as well as in pathological modes. Usage of certain values for feedback and delay parameters allows simulating the heart action when an accessory conducting pathway is present Wolff-Parkinson-White syndrome.

Fault diagnosis in a networked control system under communication constraints

Abstract: This paper considers the problem of attitude sensor fault diagnosis in a quadrotor helicopter. The proposed approach is composed of two stages. The first one is the modelling of the system attitude dynamics taking into account the induced communication constraints. Then a robust fault detection and evaluation scheme is proposed using a post-filter designed under a particular design objective. This approach is compared with previous results based on the standard Kalman filter and gives better results for sensor fault diagnosis.

An analytical iterative statistical algorithm for image reconstruction from projections

Abstract: The main purpose of the paper is to present a statistical model-based iterative approach to the problem of image reconstruction from projections. This originally formulated reconstruction algorithm is based on a maximum likelihood method with an objective adjusted to the probability distribution of measured signals obtained from an x-ray computed tomograph with parallel beam geometry. Various forms of objectives are tested. Experimental results show that an objective that is exactly tailored statistically yields the best results, and that the proposed reconstruction algorithm reconstructs an image with better quality than a conventional algorithm with convolution and back-projection.

Multiple neural network integration using a binary decision tree to improve the ECG signal recognition accuracy

Abstract: The paper presents a new system for ECG ElectroCardioGraphy signal recognition using different neural classifiers and a binary decision tree to provide one more processing stage to give the final recognition result. As the base classifiers, the three classical neural models, i.e., the MLP Multi Layer Perceptron, modified TSK Takagi-Sugeno-Kang and the SVM Support Vector Machine, will be applied. The coefficients in ECG signal decomposition using Hermite basis functions and the peak-to-peak periods of the ECG signals will be used as features for the classifiers. Numerical experiments will be performed for the recognition of different types of arrhythmia in the ECG signals taken from the MIT-BIH Massachusetts Institute of Technology and Boston's Beth Israel Hospital Arrhythmia Database. The results will be compared with individual base classifiers' performances and with other integration methods to show the high quality of the proposed solution

A RED modified weighted moving average for soft real-time application

Abstract: The popularity of TCP/IP has resulted in an increase in usage of best-effort networks for real-time communication. Much effort has been spent to ensure quality of service for soft real-time traffic over IP networks. The Internet Engineering Task Force has proposed some architecture components, such as Active Queue Management AQM. The paper investigates the influence of the weighted moving average on packet waiting time reduction for an AQM mechanism: the RED algorithm. The proposed method for computing the average queue length is based on a difference equation a recursive equation. Depending on a particular optimality criterion, proper parameters of the modified weighted moving average function can be chosen. This change will allow reducing the number of violations of timing constraints and better use of this mechanism for soft real-time transmissions. The optimization problem is solved through simulations performed in OMNeT++ and later verified experimentally on a Linux implementation

Controllability of nonlinear implicit fractional integrodifferential systems

Abstract: In this paper, we study the controllability of nonlinear fractional integrodifferential systems with implicit fractional derivative. Sufficient conditions for controllability results are obtained through the notion of the measure of noncompactness of a set and Darbo's fixed point theorem. Examples are included to verify the result.