Catálogo de publicaciones - libros

In this paper, we proposed a Chaotic Genetic Algorithm (CGA) to cluster protein interaction data to find protein complexes. Compared with other computation methods, the main advantage of this method is that it can find as many potential protein complexes as possible. Application on the Yeast genomic data highlights the efficiency of our method.

One of the key tasks in Hardware-Software Co-design is to optimally allocate, assign, and schedule resources to achieve a good balance among performance, cost, power consumption, etc. So it’s a typical multi-objective optimization problem. In this paper, a Multi-objective Q-bit coding genetic algorithm (MoQGA) is proposed to solve HW-SW co-synthesis problem in HW-SW co-design of embedded systems. The algorithm utilizes the Q-bit probability representation to model the promising area of solution space, uses multiple Q-bit models to perform search in a parallel manner, uses modified Q-bit updating strategy and quantum crossover operator to implement the efficient global search, uses an archive to preserve and select pareto optima, uses Timed Task Graph to describe the system functions, introduces multi-PRI scheduling strategy and PE slot-filling strategy to improve the time performance of system. Experimental results show that the proposed algorithm can solve the multi-objective co-synthesis problem effectively and efficiently.

Tolerance assignment is an important issue in product design and manufacturing. However, this problem is commonly formulated as nonlinear, multi-variable and high constrained model. Most of the heuristics for this problem are based on penalty function strategy which unfortunately suffers from inherent drawbacks. To overcome these drawbacks, this paper presented a new powerful tool-Particle Swarm Optimization algorithm (PSO) and meanwhile proposed a sophisticated constraints handling scheme suitable for the optimization mechanism of PSO. An example involving simultaneously assigning both design and machining tolerances based on optimum total machining cost is employed to demonstrate the efficiency and effectiveness of the proposed approach. The experimental result based on the comparison between PSO and GA show that the new PSO model is a powerful tool.

Group sightseeing has some advantages in terms of required budget and so on. Some travel agents provide package tours of group sightseeing, but participants have to follow a predetermined schedule in tour, and thus there may be no plan which perfectly satisfies the tourist’s expectation. In this paper, we formalize a problem to find group sightseeing schedules for each user from given users’ preferences and time restrictions corresponding to each destination. We also propose a Genetic Algorithm-based algorithm to solve the problem. We implemented and evaluated the method, and confirmed that our algorithm finds efficient routes for group sightseeing.

Considering fault diagnosis is a small sample problem in real chemical process industry, Support Vector Machines (SVM) is adopted as classifier to discriminate chemical process steady faults. To improve fault diagnosis performance, it is essential to reduce the dimensionality of collected data. This paper presents a modified discrete binary ant colony optimization (MDBACO) to optimize discrete combinational problems, and then further combines it with SVM to accomplishing fault feature selection. The tests of optimizing benchmark functions show the developed MDBACO is valid and effective. The fault diagnosis results and comparisons of simulations based on Tennessee Eastman Process (TEP) prove the feature selection method based on MDBACO and SVM can find the essential fault variables quickly and exactly, and greatly increases the fault diagnosis correct rates as irrelevant variables are eliminated properly.

This paper proposed an automatic recognition system of shipping container identifiers using fuzzy-based noise removal method and ART2-based self-organizing supervised learning algorithm. Generally, identifiers of a shipping container have a feature that the color of characters is black or white. Considering such a feature, in a container image, all areas excepting areas with black or white colors are regarded as noises, and areas of identifiers and noises are discriminated by using a fuzzy-based noise detection method. Noise areas are replaced with a mean pixel value of the whole image and areas of identifiers are extracted by applying the edge detection by Sobel masking operation and the vertical and horizontal block extraction in turn to the noise-removed image. Extracted areas are binarized by using the iteration binarization algorithm, and individual identifiers are extracted by applying 8-directional contour tracking method. This paper proposed an ART2-based self-organizing supervised learning algorithm for the identifier recognition, which creates nodes of the hidden layer by applying ART2 between the input and the hidden layers and improves the performance of learning by applying generalized delta learning and Delta-bar-Delta algorithm between the hidden and the output layers. Experiments using many images of shipping containers showed that the proposed identifier extraction method and the ART2-based self-organizing supervised learning algorithm are more improved compared with the methods previously proposed.

While code division multiple access (CDMA) is becoming a promising cellular communication system, the design for a CDMA cellular system configuration has posed a practical challenge in optimisation. The study in this paper proposes a hybrid estimation of distribution algorithm (HyEDA) to optimize the design of a cellular system configuration. HyEDA is a two-stage hybrid approach built on estimation of distribution algorithms (EDAs), coupled with a K-means clustering method and a simple local search algorithm. Compared with the simulated annealing method on some test instances, HyEDA has demonstrated its superiority in terms of both the overall performance in optimisation and the number of fitness evaluations required.

In this paper, a new approach for motion generation and optimization of the flexible macro-micro manipulator system is proposed based on Estimation of Distribution Algorithm (EDA). The macro-micro manipulator system is a redundant system, of which inverse kinematics remains challenging, with no generic solution to date. Here, the manipulator system configurations, or the optimal joint motions, are generated using the EDA algorithm base on Gaussian probability model. Compared with simple genetic algorithms (SGA), this approach uses fewer parameters and the time for motion optimization is remarkably reduced. The proposed approach shows excellent performance on motion generation and optimization of a flexible macro-micro manipulator system, as demonstrated by the simulation results.

A new interactive model for constructing a tactical global assets allocation through integrating fuzzy scenarios clustering- based approaches () with mean-variance () is proposed. This serves as an alternative forecasting rebalance quantitative model to the popular global assets allocation, in which the portfolio is first being observed in contrast with major asset and sub-assets classes which possess upward and downward positive co-movement phenomenon while considering the linkage of cross-market between different time-zones. In addition, fuzzy scenarios clustering would be induced into the model so as to adjust the weighting of the risk-return structural matrices. It could further enhance the efficient frontier of a portfolio as well as obtaining opportunity of excess return. By means of global major market indices as the empirical evidences, it shows that the new approach can provide a more efficient frontier for a portfolio and there would be less computational cost to solve model.

A joint source-channel coding (JSCC) scheme based on adaptive wavelet pretreatment for robust progressive image transmission over wireless fading channels using MIMO-OFDM was proposed. We pointed out that he peak signal to noise ratio (PSNR) of the reconstructed image degrades a little while the complexity decreases sharply when adding pretreatment block before DWT. On the other hand, OFDM can get good performance in the frequency selective fading channel. So in this paper, a JSCC scheme based on adaptive wavelet pretreatment for image transmission over MIMO-OFDM system was analyzed. What‘s more, the wavelet pretreatment block is loaded adaptively according to the channel condition. Simulation shows that: after adding the pretreatment block before DWT and post-treatment to the reconstructed image in good channel condition, the complexity of the SPIHT and the channel coder decreases sharply due to the compression ratio, but the PSNR of the receivedimage loses only a little; when the channel condition is getting worse ,the pretreatment block is deleted accordingly.