The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, an adaptive radial basis function (RBF) neural-networks (NNs) control algorithm is developed for a class of nonlinear affine systems based on data from controller with constrained control inputs. First, a novel nonquadratic performance index functional is introduced to overcome the nonlinear control constraints, and then the iterative adaptive optimal control algorithm is developed...
In this paper the systematic analysis and research has been made to the various learning methods of RBFNN (Radial Basis Function Neural Network). The clustering method is introduced to RBFNN. Some main methods are combined to enhance the performance in some degree. Then the improved GLVQ center learning method is put forward on the basis of ERPCL and GLVQ. Finally the GLVQ-RGLS method is the union...
Remote-sensing technology has special advantages in estimating forest carbon storage and biomass on the ground. The paper summarizes three main methods of estimating forest carbon storage by using remote-sensing technology, namely, multielement regression analytical method, artificial neural network method and mathematical modeling method. Also the advantages and disadvantages of each method are explained...
The positioning precision of multilateration system is relevant with the location of remote stations which is reflected by dilution of precision (DOP). It is a common sense that adding a new station will improve the positioning precision because of the decreasing of DOP. However, nobody has declared the quality relationship between the number of stations and DOP. In this paper, we firstly analyze...
In this paper, in order to improve the efficiency of wireless resource utilization, power control and subcarrier allocation are optimized jointly in a three-node symmetric cooperation orthogonal frequency-division multiple access uplink system. Equivalent direct channel gain is introduced to simplify relay links. Transmission mode selection strategy guarantees an optimal operation mode on the subcarriers...
Reliable and timely decision on target onset is the key of decision-based single-model techniques for maneuvering target tracking, and the performance of maneuver detector can be improved by making use of Doppler measurement extensively. In this paper, a novel approach for detecting unknown maneuver is proposed. To avoid the unresolvedness of acceleration estimator when measurement noise level is...
In order to study the ride comfort of active suspension vehicle, a detailed virtual prototype vehicle was established by using SIMPACK software, and a seven-degree freedom mathematical model for the active suspension automotive system was also built. A model reference adaptive control based on neural network was designed for active suspension system. The SIMPACK-MATLAB co-simulation method was used...
Based on contraction mapping principle, inequality technique, global exponential stability of a class of BAM neural networks with distributed delays is considered. Some sufficient conditions are derived which ensure the existence, uniqueness, global exponential stability of equilibrium points of the neural networks. Finally, the obtained results are demonstrated with a numerical example.
Aiming at the BP artificial neural network unable to auto select and optimize input variables, this paper integrates BPANN with grey relational analysis method, establishes an optimized BP artificial neural network arithmetic (GM2BPANN) which based on the grey relational analysis method. The hybrid approach has been used to forecasting the online item price. The result shows that the new model can...
Due to the inherent limitations of structure and dimensional, it is difficult to measure the surface roughness of micro-heterogeneous surface in deep hole. In this paper, the microscopic image of micro-heterogeneous surface is obtained by the long working distance lenses of digital microscopic camera, firstly. Thereafter, two artificial neural network models, which take microscopic image features...
By means of contraction mapping principle, inequality technique, global exponential stability of a class of BAM networks with delays is considered. Some sufficient conditions are obtained which ensure the existence, uniqueness, global exponential stability of equilibrium point of the networks. Finally, a numerical example is discussed in this paper to illustrate advantage of the results we derived.
This study develops and implements a SoC-based HW/SW (Hardware-Software) codesign for an intelligent diagnostic system. To improve the efficiency of the VLSI (Very Large Scale Integration) design process, the components of the intelligent diagnostic system are designed in the form of SIP (Silicon Intellectual Property) modules. The SIP modules, including the CPU module, the GPIO (General Purpose I/O)...
Accurate identification of P2P traffic makes great sense for efficient network management and reasonable utility of network resources. P2P traffic identification has two promising approaches: DPI (deep packet inspection) and DFI (deep flow inspection). The purpose of the paper is to solve how to carry out research on how to take advantage of merits of these methods, avoid their defects, and assemble...
Inductive transfer learning and semi-supervised learning are two different branches of machine learning. The former tries to reuse knowledge in labeled out-of-domain instances while the later attempts to exploit the usefulness of unlabeled in-domain instances. In this paper, we bridge the two branches by pointing out that many semi-supervised learning methods can be extended for inductive transfer...
P2P traffic has become one of the most significant portions of the network traffic. How to improve the accuracy of the traffic identification efficiently is still a difficult problem. A promising approach that has recently received some attention is traffic classification using machine learning techniques. In this paper, we propose a BP neural network algorithm for P2P traffic classification problem...
A new inductive transfer-learning algorithm called NEDRT is presented in this paper in order to improve the classification accuracy of a domain task by using the knowledge learned from labeled data generated from a different domain. NEDRT introduces a novel error function for a constructed neural network by summing a weighted squared difference between the real output and the neural network output...
To solve the problem of learning a mapping function from low-level feature space to high-level semantic space, we propose a relevance feedback scheme which is naturally conducted only on the image manifold in question rather than the total ambient space. While images are typically represented by feature vectors, the natural distance is often different from the distance induced by the ambient space...
In this paper, the establishment of the neural network model of forecasting short-term power load in an electric power grid is studied. Basing on the model, the BP algorithm for power load is explored. The research on BP network model includes determining the hidden layer number, hidden layer nodes number, training frequency and accuracy of learning rate. In this paper, we focus on that how to give...
Considering the chaotic characteristic of power system load, a method based on bee evolution modifying particle swarm optimization (BEMPSO) and chaotic neural network is presented for power system load forecasting to improve precision. In this paper, builds the chaotic neural network model and integrates bee evolution modifying with particle swarm optimization. The novel BEMPSO algorithm is proposed...
The UML activity diagrams (ADs), are lack of formal semantics in UML official specifications and therefore they cannot be performed formal system behavior analysis. This paper firstly employs the Hoare's CSP (communicating sequential processes) to formalize the behaviors of UML ADs and hence it can provide an approach to model checking UML ADs during software analysis or design stage since CSP is...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.