Search

Inferring the available bandwidth is of great importance for various network applications. In this paper, an available bandwidth measurement model is introduced based on regression model that can be used to high-speed IP network. The core of model has four parts as follows: 1) introducing the least absolute deviation method to compute regression coefficient. 2) deducing between delay and time interval. The advantages of the model are the lower cost and overhead, higher accurate result, independent clock synchronization. 3) deducing relation between sampling probability and background traffic. 4) introducing linear programming method to deal with the least absolute formula. The model is valid by simulation verification. [ABSTRACT FROM AUTHOR]

A system based on vertical linear array is brought forward. The localization model is presented, and the localization principle of vertical linear array is discussed. With range, elevation angle induced from time-delay and the measurement of azimuth angle, three dimensional equation of state and equation of measurement have been established. The localization precise according to length of baseline and precise of time-delay is simulated and analyzed. According to the observability principle of nonlinear system, the observability qualifications of the system are given. [ABSTRACT FROM AUTHOR]

Wireless sensor network should decrease the power costs of redundancy information and delay time. The technology of data aggregation can be adopted. A routing algorithm for data aggregation based on ant colony algorithm (ACAR) is presented. The main idea of this algorithm is optimization of data aggregation route by some cooperation agents called ants using the three heuristic factors about energy, distant and aggregation gain. For realizing data aggregation by positive feedback of the ants, the nodes of wireless sensor networks should not maintain the global information. The algorithm is a distributed routing algorithm and realizes data aggregation trade-off in energy and delay. The analysis and the experimental results show that the algorithm is efficient. [ABSTRACT FROM AUTHOR]

For many real-world changeable problems over time, the goal of optimization is not only to acquire an optimal solution, but also to track its progression through the search space as closely as possible. In this paper, an improved detection technique at the particle level is designed. Then, a new method of response, learning from the changing global optimum for new environments guided by population diversity, is designed. It defines response condition as well as part of particles to be reset and flying direction after a change. Then, the parabolic benchmark functions with various severities are used to test, compared with the Eberhart-PSO and APSO, and the results show the modified strategies are effective in tracking changes. [ABSTRACT FROM AUTHOR]

Sharing mechanism is introduced into particle swarm optimization. Fitness values of particles are updated to sharing fitness values. Particles with higher sharing fitness value are punished and particles with smaller sharing fitness value are remained as memory particles. Particles are updated with memory particles and clone selection when global best have not changed in some continuous generations. Population diversity is increased by this way. At the same time the particle with the best fitness value is saved. The modified algorithm can avoid the local optimization and has better search performance to multi-peak functions. The experimental results show the modified algorithm has better convergence performance than standard particle swarm optimization algorithm. [ABSTRACT FROM AUTHOR]

A class of T-S fuzzy models with parameter perturbations, input and external disturbances is studied in this paper. Both the case of matched uncertainties and the case of mismatched uncertain- ties are investigated using variable structure control technique and LMI method. Firstly, we study the problem of stabilization of T-S fuzzy systems by using a variable structure controller. Secondly, for the given reference model, we design another variable structure controller such that the system output tracks the reference system output in a finite time interval. Finally, the simulation example is employed to illustrate the validity and effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

This paper proposes a new practical optimization method applied to the economic dispatch (ED) in a power system. The proposed method is based on an improved particle swarm optimization algorithm and considers some restrict conditions of ED in a practical power system. By reinitializing them with some currently optimal values during every cycle of iteration, this proposed method can make some inactivity particles to be always within a very small area having an optimal solution. The proposed method can avoid effectively the "premature" of the classic particle swarm optimization (PSO) algorithm due to improve the cognized capacity of the classic PSO, thereby it is beneficial to obtain some optimal global solutions. The simulation results show that proposed method has some excellent characteristics of higher quality calculation precision and better computation efficiency, compared with some other PSO methods. [ABSTRACT FROM AUTHOR]

To circumvent the drawbacks in nonlinear controller designing of chemical processes, an adaptive single neuron control scheme is proposed in this paper. A class of nonlinear processes is approximated by a fuzzy neural network-based model. The key of this work is, an adaptive single neuron controller, which mimics PID controller, is considered in the proposed control scheme. Applying this result and Lyapunov stability theory, a novel-updating algorithm to adjust the parameters of the single neuron controller is presented. Simulation results illustrate the effectiveness of the proposed adaptive single neuron control scheme. [ABSTRACT FROM AUTHOR]

Object recognition technologies using PCA(principal component analysis) recognize objects by deciding representative features of objects in the model image, extracting feature vectors from objects in a image and measuring the distance between them and object representation. Given frequent recognition problems associated with the use of point-to-point distance approach, this study adopted the k-nearest neighbor technique(class-to-class) in which a group of object models of the same class is used as recognition unit for the images inputted on a continual input image. However, the robustness of recognition strategies using PCA depends on several factors, including illumination. When scene constancy is not secured due to varying illumination conditions, the learning performance the feature detector can be compromised, undermining the recognition quality. This paper proposes a new PCA recognition in which database of objects can be detected under different illuminations between input images and the model images. [ABSTRACT FROM AUTHOR]

A new decision sensor fusion model based on the fuzzy theory, which introduces fuzzy comprehensive assessment into traditional decision sensor fusion technology, is proposed in this paper. Through compare the difference between the architecture of hard decision and soft decision, the soft decision architecture had been applied. At the fusion center, the process of fusion is composed of the comprehensive operation and the global decision, and the global decision of the concerned object could be obtained by fusing the local decision of multiple sensors. In the practical application, the model has been successfully applied in the temperature fault detection and diagnosis system of Jilin Fengman Hydroelectric Simulation System. In the analyses of factual data, the performance of the system precedes that of the traditional diagnosis method. [ABSTRACT FROM AUTHOR]

A satellite cloud image is decomposed by discrete curvelet transform (DCT). In-complete Beta transform (IBT) is used to obtain non-linear gray transform curve so as to enhance the coefficients in the coarse scale in the DCT domain. GA determines optimal gray transform parameters. Information entropy is used as fitness function of GA. In order to calculate IBT in the coarse scale, fuzzy wavelet neural network (FWNN) is used to approximate the IBT. Hard-threshold method is used to reduce the noise in the high frequency sub-bands of each decomposition level respectively in the DCT domain. Inverse DCT is conducted to obtain final de-noising and enhanced image. Experimental results show that proposed algorithm can efficiently reduce the noise in the satellite cloud image while well enhancing the contrast. In performance index and visual quality, the proposed algorithm is better than traditional histogram equalization and unsharpened mask method. [ABSTRACT FROM AUTHOR]

The problem of guaranteed cost robust fuzzy filtering design for a class of time delayed systems with uncertain nonlinearities have been investigated. The nonlinear uncertain time delayed system is represented by state-space Takagi-Sugeno fuzzy model. In terms of linear matrix inequalities (LMIs), the stability of the filter error systems are analyzed by basis-dependent Lyapunov function approach. Then the sufficient conditions for the existence of the fuzzy filter are given in the form of convex optimization. The optimal upper bound of the guaranteed cost value of the filtering error systems can be solved when the LMIs are feasible. [ABSTRACT FROM AUTHOR]

This paper proposes a novel fuzzy control method based on the research of grey theory and switching algorithm in networked control systems (NCSs). A grey prediction structure is used to obtain more important information from remote sensors, and an on-line rule switching mechanism is constructed to provide an appropriate forecasting step size to the grey predictor. The overall mathematical model of this grey prediction based fuzzy controller is established. By using this method to get an appropriate positive or negative forecasting step size in NCSs, the random characteristic of some non-stationary time series from the sensors can be reduced so as to improve control performance. Experiments on a nonlinear plant via communication network show its precision and robustness is better than other traditional control methods. [ABSTRACT FROM AUTHOR]

Most of the image analysis methods focus on affective information by using color. In this paper a new indexing scheme, called FRD (Fuzzy Recognize Degree)-Clustering is proposed for affection-assisted semantic image analysis on the basis of the texture feature. A set of perceptual relevant features, used for indexing is hence introduced: directionality, contrast and coarseness. This FRD scheme allows us to retrieve images based on high-level affective concepts. Experiments with images of the nature and landscape domain demonstrate the performance of the proposed approach. [ABSTRACT FROM AUTHOR]

Abstract. The current P2P trust models suffer much from dishonest feedbacks and the trust metrics depending on subjective judgment excessively results in rough trust value. So this paper proposes a versatile trust model based on multi-dimensional trust evaluation, MDTrust. The concept of multi-dimensional trust is introduced and the approach of evaluation which uses the joint of subjective and objective evaluation is presented. We use the discrete degree of the set of evaluation similarity between peers' evaluation to measure the feedback credibility, in order to resolve the problem of collusion and bad-mouthing. Consequently, simulation and analysis shows that MDTrust can calculate the trust value effectively and reasonably and discard the malicious peers from P2P system. [ABSTRACT FROM AUTHOR]

A fuzzy multi-objective crop planning model is formulated for determining the optimal crop pattern and irrigation water resources allocation in the irrigated agriculture, in which the individual farmers and irrigation administrators put their views on three conflicting objectives: maximum of net returns, maximum of expected grain yield, and maximum of environmental returns. An interactive fuzzy multi-objective optimization (IFMOO) approach is applied to develop the sustainable crop planning for solving problems of vagueness and imprecision information related to data, model formulation and decision maker's preferences involving in multi-objective linear programming (MOLP). The methodology is illustrated in a case study of crop pattern alternatives in Xingkaihu Lake Irrigation District of northeastern China. The incorporation of these socio-economic and environmental objectives demonstrates the capability of IFMOO approach and also works suitably in water resources management by trade-off procedures. [ABSTRACT FROM AUTHOR]

A TS fuzzy model modeling method is presented in this paper. The input parameters of the TS fuzzy model are identified via fuzzy c-means clustering method and the output parameters are optimized via DNA genetic algorithm. Finally, the proposed method is applied to build the soft sensing model for the yield of acrylonitrile. Examining results demonstrate the effectiveness of this method. [ABSTRACT FROM AUTHOR]

The problem of group decision making based on intuitionistic fuzzy judgment matrix is investigated. Approaches to intuitionistic fuzzy group decision making are proposed from three different preference views. Using the operations of intuitionistic fuzzy values, the ranking method of intuitionistic fuzzy judgment matrix is given. It is illustrated by a numerical example that the approaches proposed are in accord with the rules of group decision making. [ABSTRACT FROM AUTHOR]

In this paper, a novel blind watermark extracting scheme using the Discrete Wavelet Transform (DWT) and Particle Swarm Optimization(PSO) algorithm is introduced. The watermark is embedded to the discrete multiwavelet transform (DMT) coefficients larger than some threshold values, and watermark extraction is efficiently performed via particle swarm optimization algorithm.The experimental results show that the proposed watermarking scheme results in an almost invisible difference between the watermarked image and the original image, and is robust to common image processing operations and JPEG lossy compression. [ABSTRACT FROM AUTHOR]

In this paper, we incorporate a chaotic discrete Hopfield neural network (CDHNN), as a local search scheme, into a genetic particle swarm optimization (GPSO) and develop a memetic algorithm GPSO-CDHNN for the maximum cut problem. The proposed algorithm not only performs exploration by using the population-based evolutionary search ability of the GPSO, but also performs exploitation by using the CDHNN. Simulation results show that the proposed algorithm has superior ability for maximum cut problems. [ABSTRACT FROM AUTHOR]

Estimation of distribution algorithms sample new solutions from a probability model which characterizes the distribution of promising solutions in the search space at each generation. In this paper, a modified genetic particle swarm optimization algorithm based on estimation of distribution is proposed for combinatorial optimization problems. The proposed algorithm incorporates the global statistical information collected from local best solution of all particles into the genetic particle swarm optimization. To demonstrate its performance, experiments are carried out on the knapsack problem, which is a well-known combinatorial optimization problem. The results show that the proposed algorithm has superior performance to other discrete particle swarm algorithms. [ABSTRACT FROM AUTHOR]

Ant colony optimization (ACO) is a techniqu1e for mainly optimizing the discrete optimization problem. Based on transforming the discrete binary optimization problem as a "best path" problem solved using the ant colony metaphor, a novel quantum ant colony optimization (QACO) algorithm is proposed to tackle it. Different from other ACO algorithms, Q-bit and quantum rotation gate adopted in quantum-inspired evolutionary algorithm (QEA) are introduced into QACO to represent and update the pheromone respectively. Considering the traditional rotation angle updating strategy used in QEA is improper for QACO as their updating mechanisms are different, we propose a new strategy to determine the rotation angle of QACO. The experimental results demonstrate that the proposed QACO is valid and outperforms the discrete binary particle swarm optimization algorithm and QEA in terms of the optimization ability. [ABSTRACT FROM AUTHOR]

Based on the analysis of existent evaluation methods for coaxiality errors, an intelligent evaluation method is provided in this paper. The evolutional optimum model and the calculation process are introduced in detail. According to characteristics of coaxiality error evaluation, ant colony optimization (ACO) algorithm is proposed to evaluate the minimum zone error. Compared with conventional optimum evaluation methods such as simplex search and Powell method, it can find the global optimal solution, and the precision of calculating result is very good. Then, the objective function calculation approaches for using the ACO algorithm to evaluate minimum zone error are formulated. Finally, the control experiment results evaluated by different method such as the least square, simplex search, Powell optimum methods and GA, indicate that the proposed method does provide better accuracy on coaxiality error evaluation, and it has fast convergent speed as well as using computer expediently and popularizing application easily. [ABSTRACT FROM AUTHOR]

In order to conquer the disadvantage of the traditional Immune Neural Network (INN), the paper presents INN model which is based on extenics. With matter-element analysis, the model can solve the problem that antibody identifies and memorizes antigen in immune system. The model also can make correct judgments about activation or control of nerve cell. Consequently, the structure design of INN can be optimized. And then, the new model is applied in experiment which is used for solving the problem of nonlinearity function. Based on experiment results, the model is compared with the traditional neural network. Simulation results indicate that the new model has better convergence and stability. [ABSTRACT FROM AUTHOR]

The recognition of splicing sites is an important step in gene recognition. We introduce a synthesis method for splicing sites prediction based on short sequence pattern and long sequence pattern. There are some weights in short sequence pattern and long sequence pattern. We regulate weights by simulated anneal. Applying the method to recognize both the true and false splicing sites, the result shows true positive rate and false positive rate and compared with GeneSplicer. [ABSTRACT FROM AUTHOR]

In this paper we discuss the problem of blind estimating multiple digital co-channel communication signals using an antenna array. A new approach is proposed to the problem, which is to find all the independent edges of X base on analysis of the projections of X onto coordinates. Through the simulation results of separating two blind audio signals, we can find out that blind signals could be separated with the algorithm quickly and rightly. [ABSTRACT FROM AUTHOR]

This paper proposes a method of blind separation which extracts independent signals from their convolutive mixtures. The function is acquired by modifying a network's parameters so that a cost function takes the minimum at anytime. Firstly we propose a regulation of a nonlinear principle component analysis (PCA) cost function for blind source separation of convolutive mixtures. Then by minimizing the cost function a new recursive least-squares (RLS) algorithm is developed in time domain, and we proposed two update equations for recursively computing the regularized factor. This algorithm has two stages: one is pre-whitening, the other is RLS iteration. Simulations show that our algorithm can successfully separate convolutive mixtures and has fast convergence rate. [ABSTRACT FROM AUTHOR]

An new inexact method of tangent hyperbolas (NIMTH) has been proposed recently. In NIMTH, the Newton equation and the Newton-like equation are solved respectively by one Cholesky factorization (CF) step and p preconditioned conjugate gradient (PCG) steps, periodically. The algorithm is efficient in theory. But its implementation is still restricted. In this paper, an efficient version of NIMTH is presented, in which the parameter p is independent of the complexity of the objective function, and its tensor terms can be efficiently evaluated by automatic differentiation. Further theoretical analysis and numerical experiments show that this version of NIMTH is of great competition for the middle and large scale unconstrained optimization problems. [ABSTRACT FROM AUTHOR]

We contribute two new algorithms in this paper called PImGA and PIrlEA respectively in which we construct populations online in each iteration. Every iteration process in these two algorithms does not like the normal EA and GA in which they employ the inefficient value iteration method in general, instead of, in this paper, we employ the efficient policy iteration as the computation method for searching optimal control actions or policies. Meanwhile,these algorithms also do not like general EA and GA for selection operator to get a optimal policy, instead of we make the Agent learning a good or elite policy from its parents population. The resulted policy will be as one of elements of the next population. Because this policy is obtained by taking optimal reinforcement learning algorithm and greedy policy, the new population always can be constructed by applying better policies than its parents, that is to say, the child or offspring will inherit parents' good or elite abilities. Intuitively, for a finite problem, the resulted population from simulation will accommodate the near optimal policies after a number of iterations. Our experiments show that the algorithms can work well. [ABSTRACT FROM AUTHOR]

To solve the cost control problem of outsourcing procurement inherent in the complexness and difficulty for mathematical analysis, this paper established a simulation-based cost Optimization framework. Firstly, we adopted Poisson process to simulate multi-supplier outsourcing market environment, the procurement cost expectations objective function and constraints of the model was introduced. Secondly, taking decision strategy as feedback variables, procurement cost optimization model was proposed. Thirdly, several key technologies for algorithm implement were described. Finally, based on the history demand of an aircraft model factory, numerical simulation is carried out to show that this optimization model can effectively reduce the outsourcing procurement total cost and outsourcing parts inventory fluctuation. [ABSTRACT FROM AUTHOR]

Research on evolutionary theory and statistic physics has provided computer scientists with powerful methods for designing intelligent computational algorithms, such as simulated annealing, genetic algorithm, extremal optimization, etc. These techniques have been successfully applied to a variety of scientific and engineering optimization problems. However, these methodologies only dwell on the macroscopic behaviors (i.e., the global fitness of solutions) and never unveil the microscopic mechanisms of hard computational systems. Inspired by Richard Dawkins's notion of the "selfish gene", the paper explores a novel evolutionary computational methodology for finding high-quality solutions to hard computational systems. This method, called gene optimization, successively eliminates extremely undesirable components of sub-optimal solutions based on the local fitness of genes. A near-optimal solution can be quickly obtained by the self-organized evolutionary processes of computational systems. Simulations and comparisons based on the typical NP-complete traveling salesman problem demonstrate the effectiveness and efficiency of the proposed intelligent computational method. [ABSTRACT FROM AUTHOR]

It proposed a novel dynamic structure-based neural network determination approach using orthogonal genetic algorithm with quantization in this paper. Both the parameter (the threshold of each neuron and the weight between two neurons) and the transfer function (the transfer function of each layer and the network training function) of the dynamic structure-based neural network were optimized in this proposed approach. In order to satisfy the dynamic transform of the neural network structure, the population adjustment operation was introduced into orthogonal genetic algorithm with quantization for dynamic modification of the population's dimensionality. A mathematical example was applied to evaluate this proposed approach. The experiment results suggested that this proposed approach is feasible, correct and valid. [ABSTRACT FROM AUTHOR]

Pulse Coupled Neural Networks (PCNN) is applied to image segmentation efficiently. Although the segmentation result with classical PCNN depends on the suitable concerned parameters, many experiments have shown that the segmentation result changed periodically with the calculation cyclic iteration times, N, after other parameters had been set. Therefore, how to decide the best iteration times N, is the key of applying PCNN to automated image segmentation. This paper brought forward a new edge-statistic algorithm based on calculation of connected regions, in order to automatically get the optimized value of N. An Edge-pixel Criterion was raised, and with it the algorithm calculated the valid edge pixels during the iteration process, and it meant that the maximum edge pixels were accordant with the best iteration times N, thereby the best segmentation result was achieved. The experiments show that the improved PCNN algorithm can promote the segmentation ability and has much better sensitivity than those methods based on image entropy or edge operator, and also has much stronger robustness of image noisy. [ABSTRACT FROM AUTHOR]

Artificial life computational model can be used to simulate the dynamic behavior of certain agro-ecosystem. In this paper, we present an agent-based artificial agro-ecosystem model to simulate the interaction between pest movements and trap crop physical design. The result of simulation shows that artificial agro-ecosystem model has a broad application prospect in the field of agro-ecosystem management. [ABSTRACT FROM AUTHOR]

Based on the biologic agamogenesis mechanism, the agamogenetic algorithm (AGA), which is similar to classical genetic algorithms except the agamogenesis and highly converging speed, is presented to solve the optimization problems. In order to solve the traveling salesman problem (TSP) by AGA, one agamogenetic operator R-Edge and one mutation operator NI-Dot are given by introducing the conception of the relative distance between cities. The validity of the AGA to solve the traveling salesman problem is shown by simulative experiments. [ABSTRACT FROM AUTHOR]

In Multi-Agent System, the generation of agent coalition is a key topic. It mainly researches how to generate the task-oriented optimal agent coalition. This paper proposes a novel coalition generation method based on fuzzy set theory. Defines a membership function, by which the membership degree of an agent to a coalition is calculated as a value between 0 and 1, and then the task-oriented coalition can be generated by λ matrix theory. An example is presented to illustrate the validity of the method. [ABSTRACT FROM AUTHOR]

Based on the clonal selection theory, the main mechanisms of clone are analyzed in this paper. A novel strategy algorithm based on Artificial Immune System-Immune Clonal Strategy Algorithm base on the Adaptive Mean Mutation (ICSAMM), is presented, in which the Gauss mutation in the Classical Evolutionary Strategies algorithm (CES) is replaced by the Adaptive Mean one, and the size of algorithm step can be adjusted adaptively. Compared with CES, ICSAMM is shown to be an evolutionary strategy capable of avoiding prematurity, increasing the converging speed and keeping the variety of solution in the simulation. Using the theories of Markov Chain, its convergence is proved. [ABSTRACT FROM AUTHOR]

This paper thoroughly analyzes the transfer and construction of the state-action space of the agent decision-making process, discusses the optimal strategy of agent's action selection based on Markov decision-making process, designs a neural networks model for the agent reinforcement learning, and designs the agent reinforcement learning based on neural networks. By the simulation experiment of agent's bid price in Multi-Agent Electronic Commerce System, validated the Agent Reinforcement Learning Algorithm Based on Neural Networks has very good performance and the action impending ability. [ABSTRACT FROM AUTHOR]

This paper presents a kernel-based online neuroevolutionary of augmenting topology (KO-NEAT) algorithm, which borrowing the selection mechanisms used in temporal difference (TD) algorithms and combining the kernel function approximator for individual fitness initiation. KO-NEAT can improve evolution's online performance of NEAT and learns more quickly. Empirical results in keepaway soccer problem demonstrate that KO-NEAT can substantially improve the original algorithm. [ABSTRACT FROM AUTHOR]

The particle swarm optimization (PSO) is a stochastic evolutionary computation technique based on the behavior of swarms that can be used to optimize objects with complex search spaces. However, it has been observed that its performance varies duo to the dimensionality of the object and the location of the global optimum in the search space. This paper introduces a "random" velocity boundary condition to address the problem, where the velocity boundary alters randomly to prevent the velocity of a particle from stopping on a same boundary during the evolution. Simulation results on two benchmark functions with 30 and 300 dimensionalities and three types of locations of the global optimum solutions in the search spaces have shown that with the proposed "random" velocity boundary condition, a highly competitive optimization performance can be obtained for PSO regardless of the dimensionality and the location of the global optimum solution. [ABSTRACT FROM AUTHOR]

The paper presents a method for the identification of bilinear system parameters by using an improved Genetic Algorithm. Good results could still be obtained when the system output was influenced by Gaussian noise in the simulation. By comparing with RLS and COR through a simulation experiment to a SISO bilinear system, it is found that the method can get better result than the other two methods. Through a simulation experiment to a MIMO bilinear system, the method can get reasonably good results too. These simulations show that the method is simpler and can get better results than RLS and COR. Through a simulation study to an MIMO bilinear system, good results can still be got. In the last section, the paper describes that a hybrid GA, the combination of Genetic Algorithm and nonlinear Least Square, was developed to identify bilinear system structure and parameters simultaneously. [ABSTRACT FROM AUTHOR]

In this paper, we propose the integration between Strength Pareto Evolutionary Algorithm 2 (SPEA2) with two types of coevolution concept, Competitive Coevolution (CE) and Cooperative Coevolution (CC), to solve 3 dimensional multiobjective optimization problems. The resulting algorithms are referred to as Strength Pareto Evolutionary Algorithm 2 with Competitive Coevolution (SPEA2-CE) and Strength Pareto Evolutionary Algorithm 2 with Cooperative Coevolution (SPEA2-CC). The main objective of this paper is to compare competitive against cooperative coevolution to ascertain which coevolutionary approach is preferable for multiobjective optimization. The competitive coevolution will be implemented with K-Random Opponents strategy. The performances of SPEA2-CE and SPEA2-CC for solving tri-objective problems using the DTLZ suite of test problems are presented. The results show that the cooperative approach far outperforms the competitive approach when used to augment SPEA2 for tri-objective optimization in terms of all the metrics (generational distance, spacing and coverage). [ABSTRACT FROM AUTHOR]

Theoretic researches of evolutionary algorithms have received much attention in the past few years. This paper presents the running time analysis of evolutionary algorithm for the subset sum problems. The analysis is carried out on (1+1) EA for different subset sum problems. It uses the binary representation to encode the solutions, the method "superiority of feasible point" that separate objectives and constraints to handle the constraints, and the absorbing Markov chain model to analyze the expected runtime. It is shown that the mean first hitting time of (1+1) EA for solving subset sum problems may be polynomial, exponential, or infinite. [ABSTRACT FROM AUTHOR]

In this paper, adopting the concept of autonomic nervous system, we design Self Internal Entropy Balancing System with incarnation focused on the self maintaining function, we define self type and self internal entropy as a property of system. This system checks SEG(Surviving Energy Gauge) periodically and make a balance by adjusting the treatment. In the case of the situation to survive, it keeps the system by processing the incarnation process. It is applied to knowledge network system of virtual memory and tested with sample data. [ABSTRACT FROM AUTHOR]

Many function-approaching methods such as neural network, fuzzy method are used in reinforcement learning methods for solving its huge problem space dimensions. This paper presents a novel ART2 neural network based reinforcement learning method (ART2-RL) to solve the space problem. Because of its adaptive resonance characteristic, ART2 neural network is used to process the space measurement of reinforcement learning and improve the learning speed. This paper also gives the reinforcement learning algorithm based on ART2. A simulation of path planning of mobile robot has been developed to prove the validity of ART2-RL. As the complexity of the simulation increased, the result shows that the number of collision between robot and obstacles is effectively decreased; the novel neural network model provides significant improvement in the space measurement of reinforcement learning. [ABSTRACT FROM AUTHOR]

The performance of speech recognition system is often degraded in adverse environments. Accurate Speech endpoint detection is very important for robust speech recognition. In this paper, an improved adaptive band-partitioning spectral entropy algorithm was proposed for speech endpoint detection, which utilized the weighted power spectral subtraction to boost up the signal-to-noise ratio (SNR) as well as keep the robustness. The idea of adaptive band-partitioning spectral entropy is to divide a frame into some sub-bands which the number of it could be selected adaptively, and calculate spectral entropy of them. Although it has good robustness, the accuracy degrades rapidly when the SNR are low. Therefore, the weighted power spectral subtraction is presented for reducing the spectral effects of acoustically added noise in speech. The speech recognition experiment results indicate that the recognition accuracy have improved well in adverse environments. [ABSTRACT FROM AUTHOR]

Artificial Neural Networks (ANNs) are powerful computational modeling tools, however there are still some limitations in ANNs. In this paper, we construct a new artificial neural network, which based on MAS theory and quantum computing algorithm. All nodes in this new ANN are presented as quantum computational (QC) agents, and these QC agents have learning ability via implementing reinforcement learning algorithm. This new ANN has powerful parallel-work ability and its training time is shorter than classic algorithm. Experiment results show this method is effective. [ABSTRACT FROM AUTHOR]

In recently years, fingerprint identification becomes more and more important for security. In this paper, it introduces an automatic fingerprint matching system(AFMS). It includes three stages: fingerprint classification, minutiae extraction and fingerprint matching. In the stage of fingerprint classification, a fingerprint is classified into one of the five types, Arch, Right Loop, Left Loop, Whorl and Others. In the stage of minutiae extraction, the minutiae, composed of ridge endings and bifurcations, are detected. In the stage of fingerprint matching, a matching score between two minutiae pattern is computed. Our AFMS is tested on 6 databases of fingerprint images. According to the type of top three matches, the recognition rates are excellent. The results reveal the expected performance and applicability of the system. They prove as well the availability of design methodology proposed in this system. [ABSTRACT FROM AUTHOR]

This paper introduces a mechanism for representing and recognizing case history patterns with rich internal temporal aspects. A case history is characterized as a collection of elemental cases as in conventional case-based reasoning systems, together with the corresponding temporal constraints that can be relative and/or with absolute values. A graphical representation for case histories is proposed as a directed, partially weighted and labeled simple graph. In terms of such a graphical representation, an eigen-decomposition graph matching algorithm is proposed for recognizing case history patterns. [ABSTRACT FROM AUTHOR]

A wavelet-based watermarking system is described for ownership verification of digital images. The wavelet filters used in this system are constructed by unitary transform of two-dimensional conjugate quadrature filter (CQF). Tensor-product wavelet filters are only special cases of this construction. This construction provides more ways to randomly generate perfect reconstruction filters, and will increase the difficulty for counterfeiters to gain the exact knowledge of our watermark. Furthermore, the watermark is inserted into several middle-frequency sub-bands and the existence of the watermark is asserted if any one of the correlation values is greater than a pre-determined threshold. Experimental results show that the proposed algorithm achieves invisibility, blind, and robustness to noising, sharpening, cropping etc. [ABSTRACT FROM AUTHOR]