Catálogo de publicaciones - libros

This paper presents a novel spatial domain watermarking scheme based on chaotic maps. Two chaotic maps are employed in our scheme, which is different from most of the existing chaotic watermarking methods, 1-D Logistic map is used to encrypt the watermark signal, and generalized 2-D Arnold cat map is used to encrypt the embedding position of the host image. Simulation results show that the proposed digital watermarking scheme is effective and robust to commonly used image processing operations.

Chaotic noise cancellation has potential application in both secret communication and radar target identification. To solve the problem of parameter estimation in chaotic noise, a novel radial basis function neural network (RBF-NN) -based chaotic time series data modeling method is presented in this paper. Together with the spectral analysis technique, the algorithm combines neural network’s ability to approximate any nonlinear function. Based on the flexibility of RBF-NN predictor and classical amplitude spectral analysis technique, this paper proposes a new algorithm for parameter estimation in chaotic noise. Analysis of the proposed algorithm’s principle and simulation experiments results are given out, which show the effective of the proposed method. We conclude that the study has potential application in various fields as in secret communication for narrow band interference rejection or attenuation and in radar signal processing for weak target detection and identification in sea clutter.

This Letter deals with the global exponential synchronization of a class of chaotic neural networks with time delays. Based on the the Halanay inequality technique and the Lyapunov stability theory, a delay-independent and decentralized control law is derived to ensure the exponential synchronization of the model and the simpler, less conservative and more efficient results are easy to be verified in engineering applications. Finally, an illustrative example is given to demonstrate the effectiveness of the presented synchronization scheme.

Some arguments on fuzzy neural network algorithm have been put forward, whose weights were considered as special fuzzy numbers. This paper proposes a conception of strong L-R type fuzzy number and derives a learning algorithm based on BP algorithm via level sets of strong L-R type fuzzy numbers. The special fuzzy number has been weakened to the common case. Then the range of application has been enlarged.

In this paper, a tabular reinforcement learning (RL) method is proposed based on improved fuzzy min-max (FMM) neural network. The method is named FMM-RL. The FMM neural network is used to segment the state space of the RL problem. The aim is to solve the “curse of dimensionality” problem of RL. Furthermore, the speed of convergence is improved evidently. Regions of state space serve as the hyperboxes of FMM. The minimal and maximal points of the hyperbox are used to define the state space partition boundaries. During the training of FMM neural network, the state space is partitioned via operations on hyperbox. Therefore, a favorable generalization performance of state space can be obtained. Finally, the method of this paper is applied to learn behaviors for the reactive robot. The experiment shows that the algorithm can effectively solve the problem of navigation in a complicated unknown environment.

Scaled conjugate gradient (SCG) algorithm was used to improve adaptive neuro-fuzzy inference system (ANFIS). It’s proved by applications in chaotic time-series prediction that the improved ANFIS converges with less time and fewer iterations than standard ANFIS or ANFIS improved with the Fletcher-Reeves update method. The way in which ANFIS could be improved on the basis of standard algorithm using fuzzy logic toolbox of MATLAB is dwelled on. A convenient method to realize ANFIS in TI ’s digital signal processor (DSP) TMS320C5509 is presented. Results of experiments indicate that output of ANFIS realized in DSP coincides with that in MATLAB and validate this method.

In unit steam-boiler generation, a coordinated control strategy is required to ensure a higher rate of load change without violating thermal constraints. The process is characterized by nonlinearity and uncertainty. Using of neuro-fuzzy networks (NFNs) to represent a nonlinear dynamical process is one choice. Two alternative methods of exploiting the NFNs within a generalised predictive control (GPC) framework are described. Coordinated control of steam-boiler generation using the two nonlinear GPC methods show excellent tracking and disturbance rejection results.

In this study, we introduce and investigate a genetically optimized fuzzy relation-based polynomial neural networks with the aid of information granulation (IG_gFRPNN), develop a comprehensive design methodology involving mechanisms of genetic optimization with symbolic gene type. With the aid of the information granules based on C-Means clustering, we can determine the initial location (apexes) of membership functions and initial values of polynomial function being used in the premised and consequence part of the fuzzy rules respectively. The GA-based design procedure being applied at each layer of IG_gFRPNN leads to the selection of preferred nodes with specific local characteristics (such as the number of input variables, the order of the polynomial, a collection of the specific subset of input variables, and the number of membership function) available within the network. The proposed model is contrasted with the performance of the conventional intelligent models shown in the literatures.

To cope up with the variability of defect shadows and the complexity between defect characters and classes in welding image and poor generalization of fuzzy neural network (FNN), a support vector machine (SVM)-based FNN classification algorithm for welding defect is presented. The algorithm firstly adopts supervisory fuzzy cluster to get the rules of input and output space and similarity probability is applied to calculate the importance of rules. Then the parameters and structure of FNN are determined through SVM. Finally, the FNN is trained to classify the welding defects. Simulation for recognizing defects in welding images shows the efficiency of the presented.

The control of double-inverted pendulum is one of the most difficult control problems, especially for the control of parallel-type one, because of the high complexity of control systems. To attain the prescribed accuracy in reducing control complexity, a multi-granular controller for stabilizing a double inverted pendulum system is presented based on universal logics fuzzy neural networks. It is a universal multi-granular fuzzy controller which represents the process of reaching goal at different spaces of the information granularity. When the prescribed accuracy is low, a coarse fuzzy controller can be used. As the process moves from high level to low level, the prescribed accuracy becomes higher and the information granularity to fuzzy controller becomes finer. In this controller, a rough plan is generated to reach the final goal firstly. Then, the plan is decomposed to many sub-goals which are submitted to the next lower level of hierarchy. And the more refined plans to reach these sub-goals are determined. If needed, this process of successive refinement continues until the final prescribed accuracy is obtained. In the assistance of universal logics fuzzy neural networks, more flexible structures suitable for any controlled objects can be easy obtained, which improve the performance of controllers greatly. Finally, simulation results indicate the effectiveness of the proposed controller.