Wavelet Active Media Technology and Information Processing

Preface

Contents

We define a weighted Banach space of test functions and its antidual H on a locally compact abelian group , so as to from a Gelfand triple . We obtain atomic characterization of in terms of Gabor atoms and study the boundedness properties of Gabor multipliers between these function spaces. Our Theorem 5.1 includes the corresponding results of Feichtinger and Zimmermann [FZ 98,p.l44] as particular cases. Also, we study the compactness property of Gabor multipliers on the space and , 1 ≤ p < ∞, and demonstrate that the space of all sequences of bounded variation is continuously embedded in the space of all Gabor multipliers on . This paper paves the way for the study of a number of other properties of Gabor multipliers defined on various types of Banach spaces of functions and distributions.

According to the position and movement relationship of the lining washer with respect to the outer cover of the composite bottle cap prior to their assembly and based on the structural features of the so-called extra -light, extra-thin component with unusual shape, the strategy of the lining washer is discussed for its positioning and the process of the stopping catch is kinetically presented for the lining washer filling with the guide-key with the determination of optimized oblique angle of the guide-key. The important fact, the free-falling injection of the lining washer is also studied. With the wavelet analysis, the clearance between the guide-key and the stopping catch and the falling gesture of the lining washer with respect to the free passage in the falling course based on the experiments of the two cases, leading to determination of accurate structural and kinetic parameters that secure the lining washer entering the outer-cover precisely.

In this paper we investigate a framework for bivariate non-uniform Haar wavelets. We use tensor product to define the bivariate Haar scaling function, the bivariate non-uniform Haar wavelets, and the bivariate non-uniform multiresolution analysis associated to these functions. The direct and inverse bivariate non-uniform Haar wavelet transforms are given.

Identifying the passport is important to border control police work. Wavelet decomposition and reconstruction be used to adding and detecting the digital watermark to a color image such as picture of man or woman and then made and identified the passport. Tests showed that the method we have proposed is validity to identify the passport. It can be used to avoiding the passport forgery.

In this paper, application of two-dimensional continuous wavelet transform to image processes is studied. We first show that the convolution and correlation of two continuous wavelets satisfy the required admissibility and regularity conditions, and then we derive the convolution and correlation theorem for two-dimensional continuous wavelet transform. Finally, we present numerical example showing the usefulness of applying the convolution theorem for the two-dimensional continuous wavelet transform to perform image restoration in the presence of additive noise.

Proposes a new approach of fault diagnosis for redundancy sensitive elements of strap-down inertial navigation system (SINS), based on wavelet transform. According to the time-frequency characteristics and signal edges detection of wavelet transform, the output signals of redundancy sensitive elements are decomposed in six layers by adopting db5 wavelet. Using the signals reconstructed from each layer's wavelet coefficients, the fault of redundancy sensitive elements can be diagnosed and located. Simulation results on redundancy sensitive elements show that this method not only can detect fault, fault type and occurring time, but also is sensitive to faults and robust to noise.

The paper presents how to design optimal wavelet operator that adaptively eliminate direct waves (DW) of ground penetrating radar (GPR). It is well known that there are many wavelet operators in known wavelet family so that it is important problem to select optimal wavelet from the wavelet family. Based on transcendental information of DW in GPR and under matched rule, the paper gives the method on design optimal weight wavelet operator. The experimental results by raw data of GPR show that the method on presented design of wavelet operator in paper is effective to remove DW in GPR.

The study on human visual system points out that modification on wavelet high frequency subband coefficients is hard to be perceived by human eyes. Network security technology is benefit to information security transmission. The paper put forward a security transmission framework that utilize discrete wavelet transform to transmit secrete data (such as password, etc) carried by common picture in application layer and utilize VPN gateway based on the IPSec tunnel protocol to encapsulate IP packet in network layer. It provides a kind of new choice for end-to-end security transmission.

In the paper, the notion of univariate orthogonal waveletpackets has been generalized. First, the definition for orthogonal vector-valued wavelet packets with scale = a (2 ≤ a ∈ Z) is given. A procedure for constructing the orthogonl vector-valued wavelet packets is presented. Second, the properties of thevector-valued wavelet packets are studied. Finally, a new orthogonal basis of space L²(R ,C^s) is derived from the orthogonal vector-valued wavelet packets.

This paper establishes the state equation with capability of buffer and output state equation with control item for flow-shop system by Cohens max-algebra. It also proves that the sufficient and necessary condition for the non-blocking and uniform control with no buffer is a feedback of linear state. The optimal control of minimal period on the changing order set can be transferred into the calculation for problem of trip salesman, and the optimal arrangement is a special optimal control. The important parameters of the system can be expressed directly through the man-hour matrix; accordingly it is avoidable that the difficulty of calculation is caused in the analysis for the matrix by too much dimension.

The reconstruction precision for functions or signals is one of important factors that influence the processing effect. The factors that influence the reconstruction errors of multiwavelets are proposed and discussed in the paper. These factors include the reconstruction errors of pre/post-processing methods by themselves, multiwavelets by themselves and the interaction errors of two former parts. In addition, Gibbs phenomena of multiwavelets are analyzed and discussed in the paper, and some useful conclusions are drawn in the end.

In the process of spinning, the state of machine or inconsequence of corresponding technical parameter will bring various periodic unevenness based on stochastic unevenness of yarn. Spectrum and variance-length curve of yarn signal is not only a primary method to analyze this periodic unevenness of yarn, but also a primary method to check quality of yarn and to diagnose mechanical failure. This paper provided the method to spectrum and variance-length curve of yarn based on wavelet analysis, and established the relation of them. High precision and high calculation speed is demonstrated by this method through experimental comparison and analysis.

There is strong coupling among different robotic joints, which make it become a challenging problem to design a satisfactory controller. A novel robotic joint intelligent decoupling control method is proposed based on Redial Basis Function neural network(RBF-NN), which is capable of tracking the desired trajectory with high accuracy. Dynamics of robotic joint and design and implementation of the RBF-NN-based controller is formulated. A lot of simulations are conducted, and the simulation results have shown that the control strategy is feasible and efficient.

A novel method for identification of non-linear systems is proposed to solve the problem of the estimating the unknown parameters of non-linear systems. For Haar wavelet is a set of complete, orthogonal basis and is easy to computation, so it has significant advantage in system identification and also has been applied to many other fields. Therefore, we as well utilize Haar wavelet to analyses all time functions and estimate the unknown parameters. Finally, simulation results with or without noise respectively verify the effectiveness of our method.

In this paper, a novel way of solution to wavelet equations is proposed. In terms of relationship dilation equations and wavelet equations, wavelet equations may expressed by a graph that involves dilation coefficients and dilation function. the wavelet is then solved directly by the graph. The example of two scale difference equation is given to validate the application of this way.

In this paper, we introduce matrix-valued multiresolution analysis and orthogonal matrix-valued wavelet packets. A procedure of the construction of orthogonal matrix-valued wavelet packets is presented. The property of the matrix-valued wavelet packets has been investigated. In particular, a new orthogonal basis of L²(R,C^s×s) is derived from the orthogonal matrix-valued wavelet packets.

We use Fluent software to analyze the tangential velocity, radial velocity , total pressure, and trajectory of flow of traditional unilateralism column inlet pipe and bidirectional symmetrical radial rectangular inlet pipe. And we use the unidimensional ratio of middle to outside law based on TD Hatty mathematical model to optimize the inlet pipe dimension of cyclone. The research results have provided the powerful theory support for the predecessor experiment. At the same time the research technique proposed in this paper can also be profited by researchers.

The aim of this paper is To perform single-source applications in hierarchical topology. By proposing an application level multicast approach, Hierarchical Topology Aware Grouping (HTAG), which exploits information about path overlap among members and topological hierarchy to construct overlay tree at different network layers. The results indicate the effectiveness of our approach in reducing duplicate packets and preserving available bandwidth, with reasonable delays increase. Simulations of both our protocol and the TAG over generated hierarchical topologies testify that using application level multicast can partly pad the lack of IP multicast deployment.

In this paper, the notion of orthogonal vector-valued wavelet packets of space L²(Rⁿ,C^s) is introduced. A procedure for constructing orthogonal vector-valued wavelet packets associated with dilation matrix 2I is presented. The properties of vector-valued wavelet packets in higher dimensions have been investigated. Finally, a new orthogonal basis of L²(Rⁿ,C^s) is obtained from the orthogonal vector-valued wavelet packets in higher dimensions.

Multi-path is a major impairment for mobile OFDM systems. Time variance of the mobile channel leads to a loss of sub-carrier orthogonality. In this paper, linear minimum mean square error estimator (LMMSEE) with superimposed periodic pilot for finite-impulse response (FIR) channel estimation is proposed. Theoretical analysis and computer simulation show that the proposed method are found to exhibit better performance and lower complexity than that of the LS method at SNR values of practical interest.

In this paper, the notion of vector-valued wavelet packets associated with arbitrary integer-valued dilation matrix A is introduced. First, the definition for vector-valued wavelet packets is given. A method for constructing the orthogonl vector-valued wavelet packets is presented. The properties for these multivariate vector-valued wavelet packets are investigated.

This paper puts forwards the detecting algorithm and the extracting algorithm for hidden information, which are respectively based on domain of variation and still color digital image. Efforts are made to explore the effects of the amplified information of wavelet transformation, the displaying of hidden information and its opposite process. Based on the model of the influence of information hiding on carrier properties, this paper establishes a general mathematical model for extracting hidden information and a positive transfer function and a reverse transfer function oriented to the set of “the influence of information hiding on carrier properties”. After the information is marked, analyzed and extracted, the hidden information can be restored, with satisfactory results for information recognition and analysis. As a result, it is possible to extract electronic evidence of hidden data rapidly.

This paper proposed a new watermarking embedding algorithm, which embedded the same watermarking twice for more robust. It's different from most algorithms embedded only one watermarking. Considering different frequency-bands have different anti-attack ability, we first embedded the watermarking in the low frequency band. Then choosing some suitable embedding position and combining HVS characteristics, we also embedded the watermarking in the high frequency bands. So, even one is destroyed, there will be the other one. Besides, because of embedding watermarking twice, it contains much more information, so we automatically regulated intensity of watermarking to guarantees watermarking invisible. Experimental results show that the technique is robust and invisible.

In order to predict the danger of fire, fire alarm system is designed based on single chip microcomputer AT89S51. The environment temperature can be monitored by digital temperature sensor. This paper describes the alarm search of digital temperature sensor based on many I/O ports. The structure of the scratchpad is illustrated. The Format of high and low alarm trigger TH and TL register and temperature register are described in detail. Writing scratchpad command and copying scratchpad command can keep alarm trigger register TH ,TL and configuration register unchanged in the status of power-down. The value of TH and TL can be defined by user. Amendment alarm high limit can be carried out by keyboard or the command that is sent by computer. The pivotal programs of alarm searching and alarm processing are given out. The experiment illustrates that fire alarm system can realize alarm accurately and in time.

In this paper, we study minimum-energy frame Ψ = {ψ¹, ψ², Λ, ψ^N} with compact support for L²(R), and with arbitrary integer dilation factor d, Ψ correspond to some compactly supported refinable functions, ,give a precise existence criterion of Ψ in terms of an inequality condition on the Laurent polynomial symbols of the refinable function. When Ψ does exist, d compactly supported functions are sufficient to constitute Ψ, and present a explicit formula of constructing Ψ. Numerical examples are given.

The semantic web shall enable web agents an efficient, precise, and comprehensive extraction of knowledge. Nevertheless, this new web is not likely to be adopted in the immediate future. In this paper, we describe an approach to extracted knowledge from today's non-semantic web. Our proposal is that it associates semantics web with Nature Language Parsing. This paper discusses the ontology of HowNet and the different parsing information method between NLP and semantic web. After that, the paper describes a conceptual scheme of parsing information to extract semantically-meaningful information from today's non-semantic web.

In this paper, we proposed a strong watermark algorithm for digital image protection based on wavelet. Instead of adding watermark into source image, we use it to generate a standard key to identify the copyright and ownership. The key will be sent to a third party, and there will be no modification to the source, which means this algorithm guarantees perceptual invisibility. We use a 256-level image as watermark and convert every pixel into an 8-bit binary; it has been proven that both stability and amount of embedded information were increased. Besides, taking full advantage of average compare is doubly enhanced the strength. We will illustrate its effectiveness and robustness by showing the detecting result after experimental attacks.

In this paper, we present a modified crosstalk resistant adaptive noise canceller approach. This method proceeds in two steps: First, the signal-to-crosstalk ratio (SCR) is estimated. And when the estimated SCR is less than 6dB,we suggest reconstructing a new reference signal from two original microphone signals. The second step, the basic crosstalk resistant adaptive noise canceller (CTRANC) is used to suppress noises and enhance speech. A comparative study with basic CTRANC shows the superiority of the new method in performance.

An un-blind watermarking algorithm based on discrete wavelet transform is brought forward in this article. This algorithm inherits the thought of embedded zerotree wavelet (EZM). First, use db 1 wavelet to make the image three levels decomposition, find out the wavelet tree of significant coefficients of low frequency and middle frequency and then embed watermarking used additive formula with these coefficients. Watermarking information is a black and white two-value image, the identification of intensity of embedded watermarking can use tensile factors, original image needed as watermarking detected. Experimental Results showed that it can ensure robustness of watermarking when embedded a part of watermarking information in low frequency significant coefficients and some in middle frequency

Based on wavelet transforms, an algorithm is proposed to scramble digital images. The algorithm changes both the location and the grey grade of image points. As a result, it achieves good scrambling effects and information security. When embedding digital watermarks, the fusion factors are adjusted to maximize fusion intensity using the critical threshold of human eyes for just noticeable difference (JND). Numerical experiments show good information hiding effects and high attack resistance of the hidden images, especially the ability to resist JPEG compression. The proposed algorithm can be used to protect the copyrights of digital images, and network transmission of encrypted hidden information.

This paper discusses algebraic construction methods of finite orthogonal and biorthogonal wavelet filters under vanishing moment constraints, where it gives a set of algebraic constraint equations of orthogonal wavelet filter under one vanishing moment constraint, and Daubechies wavelet filters such as dbl, db2, sym2, db3 and sym3 filters are derived from these equations as examples; and it also obtains a set of constraint equations of biorthogonal wavelets under two vanishing moment constraints, (7-5) biorthogonal wavelet filter with (4,2) vanishing moment and (5-3) biorthogonal wavelet filter with (2,2) vanishing moment are derived from these equations as examples.

A watermark algorithm based on Human Visual System (HVS) is proposed. After triple level full wavelet transform is applied on carrier image and watermark image, two wavelet coefficient trees containing 64 leave nodes are obtained. Because Fusion intensity of each sub-band image is determined based on the JND threshold of each leave node, the algorithm is self-adaptive. It is proofed by experiments that the algorithm has several advantages such as high security and stability against attacks. It can be applied on the field of digital copyright protection.

This paper presents a watermarking algorithm based on fractal coding and wavelet transform. In this method, we adopt images which are self-similar to search for its Iterated Function System(IFS) as the watermark information, considering the robust against geometric transformations of fractal code. A chaos system is utilized to control the position of watermark embedded in the DWT domain, and then append the IFS coefficients to the wavelet decomposition coefficients. The approach proves rather efficient against many kinds of distortions.

In this paper we propose a fast learning algorithm of wavelet neural networks used in approximating nonlinear functions. Geometrical structure of the network is analyzed and the method of network parameters initialization is given. In the training phase, a new improved adaptive variable learning rate is proposed for fast convergence training. It is better in approaching the nonlinear functions than the traditional BP neural network. The results of simulation indicate that the method is fast in its convergence speed and has a good approaching precision. It provides a new method in modeling nonlinear systems besides offering a beneficial reference to the identification of complex nonlinear systems.

Taking the high availability and high effectiveness of clusters into consideration, this paper puts forward a method to improve the cluster load balancing algorithm with the help of server failure detecting methods and dynamic negative feedback mechanism. Compared with other load balancing algorithms, the new algorithm greatly improves the availability of the cluster system, and improves the cluster system's overall handling capacity as well by adopting dynamic balancing.

It is difficult to construct automaton that receives sophisticate language, but we can transit simple languages to complicate languages through calculating between them. Based on the characteristic of right-line language, context free language and context sense language, this text bring up the method of constructing simple automaton to complex automaton. It is easy and utility. It is also can be used of the academic bases for proving the closing in language with FA.

Irises can be distinguished mainly due to textural or structural difference among them. In order to acquire a high recognition rate, the edge information of iris images must be employed. This paper proposed an algorithm of iris recognition based on wavelet transform. The method includes three parts: the pretreatment of the iris image, extraction of texture, matching and recognizing of iris texture. By using the abundant texture information provides by the iris images, we used the Haar wavelet transformation for texture extraction and Hamming Distant (HD) for match. This algorithm is not sensitive to illumination, noise and translation such as zoom and rotation. Experiment showed that this method achieved a recognition rate of 98.70%, it can be used in a personal identification system.

Lifting scheme is not only a fast algorithm of existing wavelet transforms, but also a tool to produce new wavelet transforms. According to the characteristic of the lifting scheme and the 9-7 wavelet, we present a new constructing of simple rational coefficient filters for lifting 9-7 wavelets in this paper. The new lifting wavelets reduce its float and integer operations, and applied the new lifting 9-7 wavelet filters to image compression, we have achieved better results.

This article indicates a new method to correct the direction of the form in the form recognition–two adjacent angles dot method. This method judges the slantwise direction of the imaginary form by the position of two adjacent angles dots, to correct the form line and characters information in the form. Also this method has been used successfully in the common form recognition system, with a small quantity of calculation, strongly resisting interference, good universal property etc.

A novel algorithm for fingerprint classification is present. Firstly, denoise and enhance the fingerprint image with wavelet technology .Thus we can do Ridge Detection and Minutiae Detection well. The flowchart of the minutiae extraction algorithm,the flowchart of fingerprint classification algorithm and experiment result of our classification algorithm on the NIST- 4 database are presented. The results of our test show the superior performance of the fingerprint classification technology based on wavelet and the great potential in the fingerprint verification.

Geometric attacks become a serious threat to most existing wavelet-based watermarking by destroying the synchronization between watermark and watermarked images. This paper presents a novel robust watermarking scheme against geometrical distortions. Using invariant centroid (IC) and a feature template in DFT domain, the synchronization can be reconstructed. The double-channel watermarking is embedded into low-frequency band in DWT domain. Experimental results prove that our watermarking scheme has good invisibility and excellent robustness against both normal attacks and geometric attacks; furthermore, this method is blind watermarking.

A novel technique for finding pathogenicity islands in genome data with multiscale neural network (MNN) is present. Firstly, decomposes different scales of genomic sequences signal into different scales of wavelets (local frequencies), and shrinking the coefficients of each scale of wavelets by means of a separate multilayer perception neural network, reconstruct the result sequence signal from the smoothed coefficients by applying the Inverse Wavelet Transform. Secondly,detect G+C patterns in genomes by comparing the result sequence with original sequences. The results on G+C patterns analysis of Dradiodurans chromosome I and N.serogroup A strain Z2491 are presented. The findings show that MNN is a powerful tool to detect pathogenicity islands in genome data.

Recognizing puny warship target in a sea is very important to anti-warship missile guidance, a method based on infrared image is suggested in this paper to realize this task. The properties of warship in infrared image are analyzed. A measure of distinguish the sky-sea line from the other objects is given. The sky-sea line is precisely located in the image. Based on the properties of warship target in the image, a rule to recognize the target is built up. A puny warship can be recognized in infrared image by this rule.

Several 1-D windows could be constructed on the direction character of sub-image in the filter window after wavelet transform. Then the adaptive filters could be obtained from the 1-D windows. With the effective weighted combination of filters in different 1-D windows, the sub-image local adaptive filters are generated. And the simulation of the adaptive filter shows it could remove the white noise of image and preserve the edge of images effectively.

The discrete wavelet analysis has become the up-to-date technology for signal processing, especially for image compression. The effect to image compression has been testified by all kinds of applications and researches. Discrete wavelet transform (DWT) represents image as a¹ sum of wavelets on the different resolution levels. Basis for wavelet transform can be composed of any function that satisfies the requirements of multiresolution analysis. It is clear that there exits a numerous selection of wavelet families in according to the choice of wavelet function. The choice of wavelet family depends on the applications. This choice in image compression application must depend on image content. In this paper, we will introduce fundamentals of wavelet based image compression. The results of image quality for different wavelet functions, compression ratios, image contents and resolution are given. So we can find comparatively better way to compress image according to the different requirements.

Time-frequency analysis is a powerful tool for the evaluation of non-stationary signals such as radar signal. Radar target motion causes time-dependent Doppler frequency in the received radar signal. This paper presents a new time-frequency analysis based on Hilbert Huang Transform (HHT) technique as a powerful tool for the inverse synthetic aperture radar (ISAR) imaging of aircraft. HHT is developed for nonlinear and non-stationary process which is just the characteristic of ISAR imaging. In this paper, the ISAR imaging model is given first. Then, the HHT-oriented time frequency analysis is newly proposed and discussed. Finally the simulation of ISAR imaging based on HHT-oriented time frequency analysis is conducted and the result demonstrates the power of this new method.

We present an entirely new fuzzy-theory-based fractal image compression algorithm. Based on Jacquin's fractal image compression algorithm, we utilize the fuzzy mathematics theory in this algorithm, and for the first time set up an image-based fuzzy integrate judgment model in which the relevant quantity of the image block is designed. The model can reflect the edge degree in the image block. When performing range-domain blocks matching, the model can significantly save the searching time. Experimental results show that the coding time can be reduced approximately by 60-70% while the PSNR of the most reconstructed images is nearly the same. Therefore this algorithm has obvious advantage in increasing the coding speed, it has wide application prospect.

Boundary Tracing method is an important preprocessing instrument for image recognition and image measurement. But the traditional reptile boundary tracing method exits a conflict between veracity and speed. This paper integrates the speediness of small window reptile and the veracity of big window reptile, and presents a new reptile with memory and alterable window for image boundary tracing. The window of the reptile is neatly altered according to the tracing requirement, and it fills the breakpoint in the rule of Minimum Distance from Pixels Center to Beeline. The method effectively solves the conflict between veracity and speed, and has well done with the problems such as trap and embranchment. The experiments for this method indicate that it has satisfying results.

In content-based audio retrieval, the audio segmentation is one of essential steps. Segmenting audio in terms to certain features will facilitate computer processing and make it easier to carry on recognition and retrieval. This paper does the wavelet transformation to audio and analyzes its approximate coefficients, then extract two wavelet-based features which are the mean approximate amplitude and the approximate zero-crossing rate. Experiments of segmentation are presented with four kinds of audio types, speech, pure music, song and silence. The results indicated that this method is not only simple but also can achieve a good effect.

In this paper, a developed segmentation method based on the wavelet transform and SOM neural network technique is proposed. This paper implements multi-resolution wavelet transform for feature extraction using the DB4 mother wavelet. The image features are extracted using statistic parameters from the multi-resolution framework. In segmentation, a SOM neural network is employed for the clustering process. Compared with other classical segmentation method, our method has the advantages of taking short segmentation time, and can be easily extended to other medicine image.

China Solar Radio Interferometer Array (CSRIA) will observe the SUN over a wide range of radio frequencies and make high spatial resolution images at many frequencies nearly simultaneously. A complete exploitation of the imaging properties of CSRIA will require an effective method to reconstruct the original image. This paper shows how multiresolution analysis and noise removal strategy with the help of wavelet transforms can be incorporated into multi-resolution image reconstruction for aperture synthesis. The method has been used to clean the dirty map generated by CSRIA, and the results testify that the method is effective and prospective.

Distinguishing one ship target wanted from other targets is very important to anti-warship missile imaging guidance; A method based on neural network is suggested in this paper to recognize ship multi-targets in infrared images. To identify exactly the target wanted, one neural network is designed for each ship target wanted. Four features are defined and are used to distinguish one ship target from the other ones. A program was designed with this method, and it was simulated in computer. The results showed this method is valid when the side of the ship faces the camera.

Under noise situations, the wavelet coefficient of noise may affect the recognition and character extraction of linear frequency modulation (LFM) signal, especially the condition of S/N (signal-noise-ratio) is lower. The wavelet transform of Gaussian noise and LFM signal is firstly analyzed. Because the modulus maximum value of wavelet coefficient of noise is reduced when increasing the scale, while its amplitude keeps invariable, we propose an adaptive-threshold method, which use the difference of wavelet coefficient between the noise and LFM signal to denoise. This method denotes a threshold according to N/S, when the wavelet coefficient is lower than the threshold, noise can be filtered. For LFM signal, so far as its frequency can bestrew the whole intermediate frequency of the receiver, then according to wavelet coefficient and its space between modulus maximum value, the LFM carrier frequency can be extracted effectively.

Mobile video streaming is a technique that the receiver could continuously view the video content while receiving the data. An interactive framework of mobile video streaming is presented in a practical way. To reduce the effects caused by the fluctuation of wireless network, an adaptive rate control method is proposed for priority process streaming with QoS control. The transmitting data rate adapts to a suitable level according to the network status. Simulation scenario has been created to help analyse the behaviour of adaptive streaming over wireless channel. Both cases with and without the proposed method were examined and the result shows the proposed modal is able to improve the overall performance in cases when the channel conditions fluctuate.

An efficient block wavelet transform method for image coding is proposed. The image is partitioned into blocks and each block is processed independently. The data of the neighboring blocks are used as overlapped part to avoid edge artifact. In each block, the spatial combinative lifting method is used for performing the wavelet transform. Compared to the global wavelet transform the memory requirement of this method is reduced to the size of the overlapped block. Furthermore, the computational operations are reduced because of the utilization of the spatial combinative lifting method. Experimental results confirm the advantages of the proposed method.

An improved coding algorithm called SPIHT coding is introduced in detail, and the algorithm can control the quantity effectively, meet the needs of disposal instantly, and provide high performance and low complexity in image compression. Experimental results show that the proposed algorithm is much efficient wavelet-based image compression algorithm.

Multiwavelets are new addition to the body of wavelet theory. In recent years, it has been a new research spot. There are many types of symmetric-antisymmetric orthonormal multiwavelets such as CL and Sa4 multiwavelets. They perfectly combine smoothness, short support, orthogonality, symmetry, ect. For the intense competition in space technology, the transmission and storage inform-ation of infrared images is very huge, the infrared images compression becomes very important. So in this paper, multiwavelet-based infrared image compression technology is researched. Experimental result shows that this algorithm is efficient in infrared image compression domain and superior to traditional methods.

This paper presents an efficient dynamic regions of interest (ROIs) transcoding scheme for JPEG2000. ROIs transcoding technique is specially value for image transmission with JPEG2000 through narrowband networks. Based on embedded block coding with optimal truncation (EBCOT), we obtain the original bitstream for the ROIs transcoding. Then, we can reorganize it without any re-encoding by using the empty packet property and Layer-Resolution-Component-Position (LRCP) progression of JPEG2000. The proposed scheme can effectively satisfy the user's needs of the dynamic definition as well as prioritization of new and existing ROIs quickly at any time during image transmission. Experimental results show that the scheme is not only highly flexible in the progressive transmission of JPEG2000 images, but also fast and easy to implement. Moreover, the proposed algorithm is fully compatible with the JPEG2000 image-coding standard, so it is also suitable for lossless transcoding of ROIs only or of the entire image.

The network statistic information should be monitored to manage the constellation networks. In this article we described a framework of on-board router traffic rate telemetry and predicting. Traffic rate statistic was gathered through SNMP packet units. The traffic rate then was processed as a time series, which was wavelet transformed with a trous algorithm. The transformed detail coefficients and final smoothed signal then were put into the Emant neural network to perform the m steps short time prediction. The prediction framework was simulated in our constellation networks emulating environment and the analysis was described in the article.

The traditional prototype of network management does not accommodate to the management of active networks, it cannot utilize the distributed computing capabilities that active networks provides. This paper introduces a pattern of active network management, studies the structure and the transmission algorithm of the active packet in the system. The paper also studies the algorithm of the active network topology discovery.

Active network provides a programmable interface to the user where users dynamically inject services into the intermediate nodes. This paper discussed the issue, which in the tradition network management exists, and has analyzed the active network management system structure and the management mechanism. The paper propose a mobile agent management mechanism based on active network, and analyze this network management system structure, and study emphasis on the system structure, the management mechanism, the management strategy.

The neural network models for intrusion detection are generally based on supervised learning algorithms with labeled data, and can't adapt to dynamic topology network. On the other hand, it is difficult to collect training data in real intrusion detection system. In this paper, a novel intrusion detection system model, based on self-organization map and mobile agent technology, which doesn't require the labeled data for training and can automatically adjust itself to transformed topology of computer network, is proposed. The experiment results show that the presented intrusion detection system model is practical and can adapt to dynamic topology in real computer network.

Binocular stereo vision is one of branches of stereo vision. This paper first introduced a model of binocular stereo vision based on computer monitor, then qualitatively analyzed the regular variations of stereo imaging; different cases were discussed according to the initial position of the image and the viewpoints of the observer. The results show that, for a line segment, there are some stereo imaging rules in four special cases; these rules may influence the stereo effect that the observer can be seen through the model.

At first there are few researches of the Chinese context formalization at present in Chinese understanding. So the character of formalization of the semantic balance and restriction, which have been summarized from the Chinese context's concepts, and the combination of the character of agent's BDI thinking state have been presented for the research of the Chinese context formalization. And the model of the semantic restriction has been set up for the single agent inner thinking state for the Chinese producing system, and the model of the semantic balance for the multi agents producing language has been set up. The two models have formed the context semantic model based on multi-agents. The related couple of semantic relation and the arithmetic of semantic composing have been presented in the thesis. The lab system of politics semanteme has been set up for the application of information security content judging and identifying. The example of “Three representative” has been tested through this model.

This article has mainly analyzed the system diagnosis PMC model symptom characteristic, obtains it satisfied the condition, proposes the new algorithm. This algorithm had the high parallelism and the development of neural network has realized the algorithm possibility with the analogous circuit, and has carried on the simulation confirmation.

Cord groove is the key contour of the cylinder used in the oil-field rig. Its complex structure makes it difficult to be processed by metal-removal operation. Based on the NC machine CK6142 with G codes, the cord groove is analyzed and machined by intelligently controlling the turning cutter's position and movement for the starting lot, the middle part-the groove, and the ending part-the climbing segment of the cylinder. Computer programs to fulfill the task are also presented. Cylinders with the cord groove machined in this way have proved to work well for the oil-field rigs in service.

When face detecting, eyes and nose are main character of face. Location of eyes and nose is very important. Eyes of people are relative rigid, face location could be based on eyes location. Combining application, first location eyes based on Hough Transform algorithm, secondly finding out location of mouth based on trigonometry of eyes and centre point of mouth. An exact location of face could be detected.

The paper proposes a music copyright protection platform. It provides services of information hiding and extracting in the form of audio digital watermarking. The information hiding algorithms of this system are designed by combining wavelet transform, data scrambling and data fusion techniques, and encapsulated.

Transient protection technology is undergoing rapid development, the main problem its practicality faced is how to recognize fault and non-fault transient. The wavelet time entropy, defined on the basis of signal wavelet analysis, can represent the time-frequency features of transient signal in many scales at the same time. Also it is robust for noise and transient amplitude. In power system, the time-frequency features of different transient signals can't be the same. This should cause the difference between wavelet time-frequency entropy distributions under multi-scale. Based on this characteristic and wavelet time entropy matrix, a classification method using these matrixes' similarity is presented in this paper. This similarity is like the one in digital picture process. The method presents standard time-entropy matrixes of all transients, and then calculates the matrixes of testing sample and entropy similarity. Make sure that the classification can reach the maximal value of similarity as possible. The transient simulations of power system fault, normal breaker switching, capacitor switching and lighting disturbance show the method is effectively in fault recognition.

This paper first introduces the development tendency and the basic characteristics of the Embedded Micro-Browser in the embedded system field, according to the present technical approaches, one kind of method which constructs the Embedded Micro-Browser (DragonFly) based-on the Linux operating system is proposed and realized, then produces the entire system modular structure and the design method. At last, it is verified that the DragonFly is one of very good software system.

The variation of flow rate and pressure head at nodes along pipeline of drop pipeline are studied in the leaking condition by the method of characteristics(MOC). The procedure is written by Visual Basic language in accordance with hydraulic transient theory to simulate and analyze the hydraulic transient property of pipeline flow so that the leakage location can be detected precisely.

The design and implementation of a Native XML database called XManager is presented in this paper. XManager covers most of the native XML database features. XManager implements a filter scheme based on XML, which makes XManager having a strong micro kernel. Based on this scheme, XManager found an open environment; any functions can be added as plug-ins. This scheme also gives XManager good performance. The experimental results show that XManager is much better than the well-known apache XML database XIndice on indexing and querying collection of XML documents.

In this paper, we talked about the three-layer system structure of PDM based on Web. In this structure, we mainly talked about the interface between database service and web service. We use XML language to realize the interface of visiting database. It is stored in XML documents of SQL lexical operating on data, and is unified managed in XML documents. It realizes the separation of the operation on database from the database. Programmer will not need to repeatedly write the operation lexical on database.

Have analyzed the problem of multi-attribute decision-making and existed methods, we bring forward methods of using wavelet neural network to built value-distribute model to improve multi-attribute decision-making, making the conform of value impersonal and accurate, also have study-self function. The application of that method in independent director producing mechanism expressed which is validity, making the produce of the independent board director more sciencetific and reasonable.

Face recognition is a rapidly growing research area due to it is contactless, most natural and most humanized. Many issues needed to do research exist in face recognition. A novel wavelet-based and SVM framework for face training and recognizing is proposed. FMT and PCA are also used for improving the accuracy and performance. In order for searching for the best performance, many experiments conducted with this method show that face recognition is largely affected by wavelet decomposing level, and at the same time by SVM type and kernel. So for high recognizing accuracy, proper methods and parameters are needed. The results prove that our scheme is an efficient method as 2D facial image recognition.

Using the high-frequency transient caused by faults to detect the faults is a new protective principle in transmission line protection; the faulted transient components include abundant faulting information. This paper combine Wavelet transform (WT), SVD(Singular Value Decomposition) and Information entropy together, then puts forward the definition and arithmetic of WTSE which can represent the composition and complexity of signals. WTSE has a good time-following characteristic so it can be used for fault detection in transmission line of power system. Simulations are carried out on a simplified transmission line model in PSCAD/EMTDC and MATLAB. The result proves that: the application in fault detection of WTSE has no relation to the fault time and won't be affected by grounding resistance, so the WTSE is an available method and has a good prospect in fault detection.

In this thesis, adaptive pole-zero collocation controller is designed aiming at the electroanalysis trough temperature character being big cubage-lag and badly asymmetry to the temperature change. At the same time, with both servo and adjustment performances taken into account, corresponding control arithmetic is put forward. The control result indicates that the electroanalysis trough temperature control has the advantages such as fast, steady, high precision etc. This system is characterized by simple structure, convenient design, good robust etc. So it provides a powerful control approach for engineering application.

In general, oil pipelines are partially or completely buried under the ground, due to many reasons such as rust and breakage, leakage accident happens frequently. Aiming at this problem, this paper utilizes wavelet transform technology to analyze and process pressure signals, and detects the leakage by using negative pressure wave method to locate the leakage position, it is effective and exact.

This paper expounds design method and hardware circuit of intelligent temperature controller based on serial chip. The watchdog circuit of controller is realized by serial chip 25045, LED display is realized by serial chip MC14489. Communication function with the computer is completed by MAX232C. The hardware and software design of A/D conversion is realized by serial chip MAX187. For the D/A output, not only the hardware connection and software program are provided, but also the time schedule of D/A converting circuit is analyzed. Compared with the past design, All of the chips which are used by the intelligent temperature controller are serial chips. The advantage is that it has small volume and saves I/O resource of CPU. The system has high reliability and stable performance and it is used expediently.

Many genetic disease which brought out by genetic reason basically all the same intermarriage which mainly caused by distort heredity and abnormal evolution can be cured by wavelet analysis theory and technology instruments turning against genetic disease and preventing recessive genetic pathology from coming into being and eliminate some multi polar hereditary morbid.

It is dissatisfactory to implement wavelet threshold to de-noise only with wavelet coefficient magnitudes. In this paper two novel methods of wavelet threshold de-noising based on locale variances are presented. They use the wavelet coefficient magnitudes as well as the local variances, which denote the clustering properties of wavelet coefficients preferably. Some simulation results, obtained from recovering a signal embedded in the additive white noise in different signal-to-noise (SNR) settings, show the potential and effectiveness of the proposed methods.

In this paper, a new hybrid algorithm which combines ant colony system (ACS) with genetic algorithm (GA) is proposed. The proposed algorithm adds GA to ACS' every iteration. ACS' convergence speed is quickened due to introducing GA's advantage of whole quick convergence. And GA's mutation mechanism improves the capability of ACS to avoid being premature. Furthermore, this paper shows the new algorithm's validity to solve such a combinatorial optimization problem as TSP with the experimental results.

In this paper, a new kind of automated fault diagnosis algorithm is proposed. It is a new data-mining algorithm with ontology-based. General data mining algorithm can find rules in database, but they are described by low-level concepts, many of them are not interesting. In this paper, we use ontologies provide a mechanism, in data mining, by which domain specific knowledge many be included to aid the discovery process and can find multi- level classification rules, the rules are described by high-level concepts. So it can provide more interesting rules for enterprise. Besides, a study case is given to explain the practical application with the fault diagnosis bases on ontology, ensuing provides more interesting rules to make decision.

Immune algorithm can identify and maintain multiple optima of a multimodal function. But immune algorithm only uses part of excellent characteristics of immune system, and it has some shortcomings such as slow convergence rate and weak diversity. So an improved immune method is introduced in this paper. It develops the original algorithm by use of negative selection that can speed up the convergence and the mechanism of hierarchy that can improve the veracity of identifying multiple optimal. Furthermore, the experimental results show the validity of this new method to solve the multi-peak-value function optimization problem.

We use a wavelet based estimator for the parameter of long-range dependent process to analyze the scaling nature of internet traffic. Wherever and whenever these internet traffic are collected from, they exhibit long-range dependent properties. And the range of Hurst parameter is [0.5,1].

In the conventional synchronization schemes based on pilot and the guard intervals, frame synchronization couldn't been achieved for DVB-T with high mobile speed. Two novel frame synchronization schemes based the training sequence/Barker code with very lower power were proposed and compared for European DVB-T 2K systems to improve the mobile performances. Both the estimators here designed to achieve better frame synchronization in Rayleigh fading channels with mobile speed 150 km/h, especially where the second scheme could achieve 99.9% correct frame synchronization probability with SNR 5dB. Simulation results also showed that the superimposed training sequence/Barker code had a significantly smaller effect on system performance. Obviously, the proposed schemes are much better than the conventional schemes.

Image after transforming in multi-layer wavelet, every adjacent coefficient in each sub-band has dependence. And neural network have good nonlinear mapping ability can be used to set up the pertinence model, and then select several position from the model to embedding watermark. In addition, with trained well neural network it can implement blind detect when detect the watermark. The experiment verifies, the watermark image has stronger resisting ability of attacking. It's a kind of good blind watermark detect algorithms.

A new method for parameter design, where the cost is influenced remarkably by the parameter central value, had been put forward by me in 1996. In this paper, the quality loss function is improved based on it so that a synthetic method, in which parameter design and tolerance design can be made in progress at the same time, can be constructed. The synthetic method can theoretically achieve the minimum quality loss value and it is verified by a calculating case of mechanical design that the synthetic method can obtain a smaller quality loss value than Taguchi's classic method

In this paper, a wavelet-based initialization method is presented for codebook generation in vector quantization image compression. The discrete wavelet transform that keeps the maximum characteristics of blocks is put in use to produce the initial codebook of LBG algorithm and, in the meantime, to hasten the convergence rate of the final codebook. Experimental results showed that the proposed algorithm performs well when compared to the previous codebook generation method, and its final codebook is provided with representative codevectors which lead to high output image quality.

The unstable equilibrium points of a new class of simple 3-neuron chaotic cellular neural network can be stabilized very easily via linear feedback control. The stability condition is obtained. Simulations results indicate that this method in the paper is available.

In this paper, a new blind equalization algorithm based on fuzzy neural network (FNN) is proposed. It makes use of blind estimation (BE) and FNN classifier to equalize. Firstly BE algorithm is used to identify the channel character, the signals are rebuilt by deconvolution, and then the signals are classified by FNN classifier. This algorithm has the merits than the foregoing neural network algorithm, such as faster convergence speed, smaller residual error, lower bit error rate (BER), etc. The validity is proved by simulations.

In the paper, Some solvability conditions of Periodic Solutions are obtained for a class of Second-order Hamiltonian System by the mini-max methods.

It is a good choice for OFDM systems using superimposed training sequences to achieve synchronization, channel estimation and equalization, etc. However, many problems such as superimposed training power distribution are still open for us to solve. The power distribution scheme and the synchronization scheme with superimposed training are proposed to solve these problems. Theory analysis and simulation show that these schemes are useful for OFDM systems with superimposed training to improve the performances.

According to Bezout's theorem, the Coiflet solutions are a finite set for a given order of the vanishing moments. The Coiflets given by Daubechie, et al, are only one particular solution for each order. For various order and offset, we have constructed a number of Coiflets, of which some filter coefficients and plots are given. The continuation method is used instead of Newton's method. The continuation is a more effective method than Newton's one.

Traffic flow prediction is one of important problem in ITS. Now there are many kinds of algorithm which are based on accuracy and real-time of prediction. In this paper, based on local and add-weight local prediction method of chaos time series, a new kind of prediction algorithm is presented by mix together with their advantage. A unified prediction model can be used for traffic flow forecasting, practical for both condition of local and add-weight local prediction method, but the complexity is not high. Results in experiments shows that it is high accuracy and can be used in traffic flow forecasting.

The two-machine flow shop scheduling problem is considered with linear processing time and chains constraint in order to minimize the makespan. Under the simple linear deteriorating, nonnegative start and stop lag deteriorating rates are introduced so as to combine jobs and construct composite jobs satisfying precedence relations. Finally, the optimal polynomial algorithm for the two-machine flow shop scheduling problem with chains constraint is given.

The simulation model of four stroke gasoline engine is established first according to the four stroke gasoline engine's operation principle in this paper. Then the simulation result and the result contrasting to the experiment are given out and the gasoline engine speed-governing system model is established based on the valve control, and the simulation result is also given out. Guided by the simulation model and simulation result, the design of software and hardware for electronic Speed-governing System is carried out, and the result of the experiment is given out.

In this paper, an efficient data compression scheme using wavelet tree structure and improved LBG algorithm is proposed. First, the image is decomposition by wavelet transform with selected wavelet basis. Second, the wavelet tree structure can be made of the decorrelation coefficient. In general, we use the traditional training algorithm for vector quantization such as the LBG algorithm, here, we presented a novel training algorithm for vector quantization in which the convergence of the entropy sequence of each region sequence is used as the condition of the end of the algorithm. Compared with the LBG algorithm, it is simple, fast and easy to be controlled. Finally, We test the performance of the algorithm by image Lena. The result shows they can obtain the better compression rate, but the running time of it is at most one second of LBG.

Some methods of similarity measure between vague values (sets) in existence are researched, and some errors and limitations in some former result are analyzed. A general definition of similarity measure between values (sets) is presented, and two ameliorative similarity measures between values (sets) are presented. Finally, the ameliorative similarity measures between values (sets) are validated by experiment data.

No abstract received.

In this paper, a new multi-input wavelet network with one hidden layer and single output is proposed. The structure of this network is similar to that of the radial basis function (RBF) network, except that the radial basis functions are replaced by 1-D orthonormal wavelets. In the learning algorithm, training data and forecasting data are important factor in wavelets selecting. The weights are determined by linear equations. The learning error of network is zero when the forecasting data is a compact subset. This kind of network is able to approximate an arbitrary nonlinear function in multi-dimensional space and has robustness is demonstrated through theoretical analysis.

Resource sharing is an important issue for almost all applications. In this paper, we study communication link sharing strategies for application level scheduler on grids as well as heterogeneous and distributed characteristics. We assume that the link is shared via time-sharing or proportional-sharing. We build solution to solve this problem by linear programming, in rational numbers, which can be solved in polynomial time. Our preliminary results show that time-sharing scheme may achieve better performance than that of proportional-sharing.

On the basis of the thresholds of legible symbols or basic elements the authors define the limit circle, and then authors propose the mathematical definition of scale, non-scale and semi-scale cartographical symbols based on the theory of internal, external point and the orbit. According to the cartographic symbol characters of point set and the topology concepts of neighborhood, boundary and connected set etc, the authors propose the mathematical definition respectively for point, line and area cartographic symbol. Form the definitions the authors draw the related rules between scale, non-scale and semi-scale cartographical symbols and point, line and area cartographic symbols. In the end expounded dependence relations are also pointed out that figure and ground play the whole function in the transmission of geographic information.

Based on minimum the output energy (MOE) , this paper constructs a new object function from combining penalty function with Lagrange function, and proposed an improved blind multiuser detection algorithm based on Lagrange neural network. The theory analysis and simulation shows that the improved algorithm has lower computational complexity, faster convergence speed and lower bit error rate.

In order to simplify the encoding process in vector quantization (VQ) image compression, a novel fast codebook search algorithm is presented in this paper. This algorithm integrates the technique of principle component analysis (PCA) with discrete wavelet transform (DWT) to quickly locate a reasonable initial codeword. Due to the smooth property of natural images, in addition, an adaptive search scope can be chosen by referring to the neighboring wavelet amplitudes and further used for the image block to be encoded. Experimental results showed that the proposed algorithm can indeed help providing a fast VQ encoder that only requires a few candidates for full comparison.

A classified method based on wavelet transform and fuzzy c-means clustering was proposed in this paper, which classified the apple grade from the parameters of color, shape and diameter. The results show that compared with other edge detection methods, the wavelet transforms approach get a better image edge map. The experimental results show that classifier based fuzzy c-means clustering is better than traditional learning vector quantization (LVQ) networks classifier

HF OTH radar can accomplish surveillance over the horizon. But backscatter signal that is output from receiver may contain strong noise and interference background. Especially, these impulsive interferences such as lightning may inject transient and great power into the receiver, arise the noise floorin the range-Doppler map and decrease SNR. So these powerful interferences must be suppressed before target tracking. Many solutions are present for these excisions. Meanwhile thestrong ocean clutter must also be firstly suppressed. Different from these approaches at present, this paper proposes a novel algorithm of direct detection without in advance depressing clutter and these impulsive interferences. The main idea is focused on the different local regularity between target return, interference and noise. Herein the combination of Lipschitz exponent and CWT is NOT for noise suppression and interference detection in popular techniques any more, but used for the local regularity estimation and signal classification.

In order to overcome the disadvantage of traditional methods of fault features extraction, and to realize the on-line and intelligent fault diagnosis, a new method of feature extraction based on the lifting wavelet packet transform is presented, with which fault feature factors were extracted form Four typical running states of gearbox. The results show that the method of fault feature extraction based on the lifting WPT can reduce the need of time and memory greatly, and it is very fit to the real-time fault diagnosis system.

Emitter recognition problem plays an important role in electronical countermeasure intelligence processing. Practically, emitter information detected by multisensor system takes on uncertainty, illegibility and contradiction. Rough set theory is a relatively new soft computing tool to deal with vagueness and uncertainty, and is regarded as a field of leading edge. In order to solve emitter recognition problem, rough set theory is introduced. A new emitter recognition algorithm is presented in this paper. And decision-making method is discussed. At last, application example is given, which demonstrates this new method is accurate and effective. Moreover, computer simulation of recognizing the emitter class is selected, and compared with classical statistical recognition algorithm through simulation. Experiments results demonstrate the excellent performance of this new recognition method.

Efficient partitioning of large data sets into homogenous clusters is a fundamental problem in data mining. The standard hierarchical clustering methods provide no solution for this problem due to their computational inefficiency. The k-means based methods are promising for their efficiency in processing large data sets. However, their use is often limited to numeric data. But mixed data with numeric data and categorical data are extensive used. K-prototypes algorithm is based on the k-means paradigm but removes the numeric data limitation whilst preserving its efficiency. This paper researched the k-prototype algorithm and simulated it with mixed data .The simulation result express k-prototype algorithm is effective for clustering large data sets because it's time cost is linear increase by degrees.

In this paper, we introduce a new class of nonlinear variational inclusions with H-monotone mappings. By using the resolvent operator method associated with H-monotone mappings in Hilbert spaces, we prove the existence and uniqueness of solution for this class of nonlinear variational inclusions. We also construct a new Ishikawa type iterative algorithm for finding approximate the solution of this nonlinear variational inclusion and discuss the convergence and stability of sequence of iterates generated by the algorithm. The present results improve and extend many known results in the literature.

In this paper, we report on the application of wavelet-based FDTD in numerical simulation of erbium-ytterbium co-doped waveguide amplifier (EYCDWA) using multi resolution analysis (MRA). The difference formulation for multiresolution time domain(MRTD) based on Daubechies wavelets is derived, and the gain characteristics of EYCDWA is simulated by MRTD. It demonstrates that this new algorithm is valid and shows good agreement with experimental results.

Elliptic scalar multiplication is the most cost operation of the elliptic curve cryptosystems. The fast implementation of Elliptic Curve Cryptography mainly depends on the fast computation of elliptic scalar multiplication kP. This paper analysis and compare the complexity of the “add and double methods” and Montgomery methods. We show that the Montgomery methods has more cost than the “add-and-double” method if the ratio of the cost of field inversion and field multiplication is more than 2 under the affine coordinate representations without concern the cost of field adding and field squaring in GF(2^m). In fact, the ration of I and M is more than 7 as the size of binary field is not less than 128, which means that the algorithm 3.1 outperforms the algorithm 3.2.

In this paper, we introduce and study a new system of set-valued variational inclusions involving H-monotone mappings in Hilbert spaces. By using the resolvent operator method associated with H-monotone mappings, we construct a new iterative algorithm approximating the solution of this system of set-valued variational inclusion and discuss the convergence analysis of the algorithm.

In this paper, a new class of general nonlinear quasi-variational inclusions involving m–η – accretive operators in Banach spaces is introduced and studied. And a new iterative algorithm for this class of general nonlinear quasi-variational inclusions and its convergence result are established.

In this paper, we introduce the notion of vector-valued multiresolutionanalysis and biorthogonal vector-valued wavelets. A constructive method of a class of compactly supported biorthogonal vector-valued wavelets with two-scale is proposed by using multiresolution analysis and matrix theory.

The purpose of this paper is to introduce a new class of general nonlinear variational inclusions with H - accretive operators in Banach spaces and prove its equivalence with a class of fixed point problems by making use of the properties of H - accretive operator, we also prove the existence of solution for this general nonlinear variational inclusions and the convergence of iterative sequences generated by new perturbed algorithm. The results presented in this paper improve and generalize some recent results in this field.

To survive kinds of image manipulation, embedding the watermark into only one frequency subband usually can not provide a satisfied performance. In this paper, we proposed a new algorithm to embed watermark into still gray image. We embed watermark into both the low frequency coefficients and the high frequency coefficients with different ratio. The result of our experiment shows a better robustness of the watermark both in lowpass environment and in highpass environment.

It is pointed out that for modified Wien-bridge chaotic circuit, one is able to realize the chaos synchronization between the chaotic circuit and its slaved system via Open-Plus-closed-Loop (OPCL) method. Constant Hurwitz matrix in this OPCL method is obtained by theory method, and some numerical calculation results are given. The result in this paper indicates that the OPCL method can be applied to piecewise-linear chaos system.

One of the main obstacles on reliable high-speed underwater acoustic communications through severely band-limited underwater acoustic channels is the intersymbol interference (ISI) caused by the multipath effect. Channel equalization technique is an effective way to combat ISI. One of the main disadvantages of the most commonly used constant modulus algorithm (CMA) is its slow convergence rate. In order to speed up the convergence rate, a new modified approach named decorrelation based constant modulus algorithm (DBCMA) for eliminate the underwater channel resulting multipath effect is proposed, and the computer simulation results confirm the effectiveness of the proposed algorithm.

A multi-watermark method is presented in this paper. Multiwavelet is the new development in the wavelet theory, which has better characteristics than wavelet. Make the best of the multi-channel character; multi-watermark based on SVD (singular value decomposition) is embedded in the multiwavelet subbands. Experimental results show this method is efficient for embedding multiple different grayscale image watermarks at the same time.

According to Bezout theorem,Daubechies wavelets are a finite set for a given order N of vanishing moments. Available Daubechies wavelets are only one particular solution for each order N. We have derived all solutions of Daubechies wavelet for each order N from 2 to 10.The continuation method is used.

Adaptive filters are important application in the signal processing field. This paper discusses the adaptive local noise removal algorithm proposed by Professor Rafael C. Gonzalez and points out its shortcomings. To improve the algorithm, a new method, geometric mean based adaptive local noise removal algorithm, has been proposed. The simulation results indicate that the new algorithm is more satisfactory. The mean square error e_mse is reduced by 1/4. The signals to noise ratios (i.e., SNR, SNR_m, PSRN ) are raised by 1/10.This algorithm has been shown promise for applications.

In this paper, a novel algorithm for numerical solutions of American option models is presented in terms of the precision integration and wavelet. We discretize the space variable employing quasi–wavelet appropriately so that the resulting semi–discrete system of equations is a system of ordinary differential equation, and then we apply precise integration method to the semi-discrete system. Numerical results show that the proposed algorithm is practicable and efficient.

Letter fusion is one of the most important stages in letter recognition. Tang Yuan Yan first used wavelet transform (WT) in this area. However, multiwavelet possesses desirable properties in compassion with scale-valued wavelet. In this paper, a new algorithm for letter fusion is presented based on multiwavelet transform (MWT). We apply various wavelets, in particularly, the analysis-ready multiwavlets (Armlet) which was constructed by Lian jian-ao recently to our experiment. Experimental results show that balanced multiwavlet transform (BMWT) is more effective than MWT on merging letters, BMWT of higher balanced order indicates larger entropy and Armlet has better performance that approximates to the balanced wavelet.

A new blind image watermarking algorithm robust against both geometric attack and JPEG compression is proposed. The proposed discrete wavelet transform-fast-Fourier transform(DWT-FFT) composite watermarking scheme embeds a gray image in subband in the DWT domain, and embeds a template in magnitude spectrum in FFT domain. Experimental results have demonstrated that the watermark is robust against both affine transformations and JPEG compression well.

Video coding technology always is a research hot and difficult spot. Hybrid coding algorithm of MC/DCT is used in H.26x and MPEG-x video compression standard, which always introduces blocking effect because of block DCT. In order to overcome this drawback, hybrid coding algorithm of MC/DWT is applied in this algorithm. Further research shows that traditional wavelet is not always superior to DCT in video coding algorithm. And multiwavelet video coding algorithm based on block motion compensation is studied in this paper. Experimental result shows that this algorithm is superior to the hybrid algorithm of MC/DCT and MC/DWT.

This paper presents an improved dual-watermarking scheme for two purposes: one is to protect the copyright of an image with a robust watermark, and the other is to verify integrity of ROI (region of interest) with a fragile watermark. These two watermarks are embedded in different domains with different techniques. The advanced robust watermarking correlating the watermark information sequence with the watermark position sequence is embedded into coefficients of ROB (region of backgrounds) after DWT. Then the fragile-watermarking is embedded into the LSB (least significant bit) of ROI in spatial domain. Both theoretic analysis and experiment show that our proposed dual-watermarking scheme is effective to affirm copyright and verify content integrity simultaneity; furthermore, robustness and security of the robust watermark are greatly improved with associated-sequences technique.

Drawing on the rich theory of wavelets, a new method for vehicle license plate image denoising with edge preservation and enhancement is proposed in this paper, based on image multi-resolution decomposition by a redundant wavelet transform. As the experimental results show, a noticeable enhancement of number plate image quality(less noise without loss of important details) can be obtained by using this method. Furthermore, a good base for future segmentation or recognition is built.

A new method for mixing watermark generation is proposed in this paper. Based on principal component analysis, a multi-resolution transformation coder in bipolar domain is presented. Spatial recomposition is given for reassembling transformed signal. The generated watermark is provided with pseudorandom distributed structure and flexible embedding pattern. Experimental results show that the proposed method is effective and has good robustness to common image signal processing attacks.

We study the precise positioning of IC (integrate circuit) chip solder bridge in this paper. Image edge detection algorithm based on wavelet transform is a promising and effective algorithm among many edge detection algorithms. In this paper the principle of wavelet transform is applied to precise positioning of IC chip solder bridge, a method of small goal detection algorithm is proposed, this detection method can accurately detect and position the tiny solder bridge from the single frame image, and it is practical and effective. Comparing to searching the whole image, it greatly enhances the detection speed.

In this paper, we develop a new approach to mammographic image enhancement and denoising based on redundant wavelet transform. The first part deals with the method of mammographic enhancement. Firstly, we enhance the image using a fast and effective enhancement method by the local standard deviation values, and stretch/compress factitiously the gray interval between neighbor pixel for different gray value according to the human visual curve in Figure 1 considering the character of the human visual system, so as to distribute the redundant gray of some gray value to eye to the wanting fields. The method boosts up the local regions to different degrees rationally. The second part deals with the multiscale image processing for image denosing. The image is decomposed into several subbands by redundant wavelet transform, and then is reconstructed by different weights.

A wavelet-based watermarking algorithm is proposed in this paper. By using J M Shapiro's view [2] about zero-tree, we embed four watermarks “reversed” to each other in different levels of the zero-tree. During the embedding of the watermarks, the Human Visual Model (HVM) is introduced to maximize the hiding capacity. Due to HVM and the characteristic of “reverse” of the four watermarks, although the magnitude of the data which are embedded in the host image is large, we can't feel any distortion of the watermarked image. Moreover, we develop a new complementary detecting technology, using which five watermarks are detected in the process of detecting; they may complement each other to reduce the detecting error. The experimental results show that this method can resist various attacks.

Image communication with high speed and low bit error rate has attracted almost all the researcher's attention in wireless communication. Meanwhile, OFDM is regarded as a key technology of next generation wireless mobile communication. So, in this paper, we discussed image transmission over Rayleigh fading channel with OFDM modulation. Results show that OFDM can achieve high data transmission speed, this is very useful for image transmission which need high transmission speed to meet its large data quantity. Moreover, OFDM can achieve reliable and high efficiency transmission of image, and, it can resist frequency-selective and inter-symbol interference. All the results indicate that OFDM modulation is very suitable for image transmission and the distort image can be recovered by OFDM modulation at 12dB.

We don't consider the burst of the traffic stream in traditional network, it is short-range dependence. The recent research has shown the existence of self-similarity and long-range dependence in real network traffic. First of all, we give several familiar definitions of self-similarity (also called long-range dependence), describing the stochastic process's characteristics in mathematics and physics, and then we will discuss how to identify a sequence is self-similarity (long-dependence) or not and how to identify its magnitude in quantity. In this paper, we produce the traffic source with the Pareto distribution (one of the heavy-trail distribution) by the ON/OFF model. Subsequently, we estimate the H parameter using V-T and wavelet methods. As the V-T method, we emphasize on the exactness and the operation time. As the Wavelet method, we analyze the effect that the parameters of wavelet method put on the result and introduce the optimal scheme for choosing the parameters. Base on the optimal scheme, we gain very good result. The result of simulation shows that the estimated parameters correctly describe the characteristics of self-similar traffic stream. Whether exactness or the operation time, Wavelet method performs better. At the same time, we study the effect that the vanishing moment of wavelet put on the result by simulation. We gain the applicable scope of wavelet method.

The Set Partitioning into Hierarchical Trees algorithm for wavelet-based image coding allow fully scalable transmission. In this paper, two methods were proposed to improve the coding efficiency of the SPIHT algorithm. In the first, the predictable coding redundancies in the algorithm are removed; on the other hand, the second method takes advantage of the characteristic that different subbands have different magnitude distributions. Finally, experiment results are also given to demonstrate the effectiveness of the proposed methods.

Acupoints are important in acupuncture. But research on acupoints signals and their processing is seldom met. In this paper, we acquire acupoints signals from three acupoints of “Quchi (LI 11)”, “Hegu (LI 4)” and “Lieque (LU 7)”. Using the wavelet transform, we de-noise the signals. The de-noising steps include decomposition, thresholding, and reconstructing of the signals. The results show that noise signals are successfully de-noised from the mixture signals of acupoints.

This paper proposes a novel blind digital watermarking scheme for grayscale image based on Integer Wavelet Transform (IWT). First of all, the watermark grayscale image is scrambled by Arnold Transform, and compressed by IWT, then transformed into binary sequence of watermark. At the same time, N coefficients are selected from the low frequency LL subband coefficients of IWT to be the embedding domain and their five lower bits are used to embed the watermark. As to the detection algorithm, a novel difference detection algorithm is proposed to take the place of traditional position detection algorithm. Experimental results show that the algorithm has the following characters: high capacity, imperceptibiliry and robustness against some common image processing operations, JPEG compression, cropping and geometric transformations.

Wavelet transform is a kind of method of signal analyse based on time–scale. The method has a specialty of multi-resolving power, and that can represent the local characters of signal in time field and frequency field. Both time window and frequency window can change. Time–frequency localized in wavelet.

Some key techniques for CBIR, including visual features description and extraction, similarity/distance measures, system performance evaluation, is pictured. The principle of CBIR is to make a similarity comparison and distance calculation between features of query image and images in database. Color, texture, shape, and spatial information etc., general visual features most widely used in CBIR are discussed. Various ways to calculate the similarity distances between visual features are presented. Furthermore, QBIC is introduced as a prototype system.

It's important to filter off noise component and keep the figure's geometry configuration during the disposal of the image. Smooth filter algorithm of noise image is used in this paper to eliminate noise commendably and intensify the image edge without blurring the image edge and damaging the detail inside the image.

Information hiding technology is widely used in various fields, such as secret communication, digital watermark, forgery detection, etc. The paper proposes a method of information hiding based on wavelet transform, the procedures for hiding and extracting information are also presented in this paper. The method has good robustness and can resist some attacks such as lossy compression, noise, etc.

This paper addresses the problem of noise reduction of image based on wavelet transform under non-stationary environments. The conventional methods of noise reduction are only efficient for stationary environments and have much image residual noise. The method proposed in this paper is based on the noise distributing characteristics and the relationship between scale and the wavelet modulus maxima. Using this method, we can obtain the position and coefficient value of the noise in the wavelet domain through iterations. Experiment results confirm that the algorithm keeps more details and has higher peak signal to noise ratio (PSNR) than traditional methods, and it is effective to reduce the noise under stationary and non-stationary environments.

This paper proposes a multi-agent data process method based on Rough Set in grid environment. It can effectively process data by Bayesian classification based on attributes reduction in Rough Set and dispatch a collection of agents to coordinate a user job over grid computing. Our method can well resolve the main problems about data process that exist in grid. In this paper, we firstly illuminate the Bayesian classification based on Rough Set. Then we detail multi-agent's framework and scheduling mechanism. We also illuminate the agent's implementation based on Java. Finally we represent the experiment result by comparing our method with the naïve grid.

Nowadays, with more industrial control networks connected to IT network, or more IT technologies applied to industrial automation systems, their security becomes a more focused issue. After introducing the general industrial control network architecture and analyzing the threats against it, we present a DMZ model based on security zone for the first time, which can effectively prevent control networks from malicious attacks.

An extended model for coupled networks considering three possible links, i.e. rewiring links, direct links, and cross links, is proposed in this paper. Following the establishment of the master equations of degree distributions, the exact asymptotic solutions in power law form and their corresponding exponents are obtained. It is indicated that the minimal model used can describe the acquaintance webs well. The results also show that more other known consequences can be inferred just by tuning the parameters properly.

Follow the trend of the combination of financial and non-financial measures of strategic performance measurement system. A performance measurement index system for investment decision-making was designed which consisted of six-measured, including financial condition, theological innovation, customers, business process, latent development capability, and social and environment protection. This index system is comprehensive flirtation combined by Benchmarking and Principal Component analysis. Wavelet network being better than neural network for solving nonlinear problems overcoming shortcomings of neural network's constringency with slow velocity and easily entering into local minimum, the performance evaluation model for investment decision-making based on wavelet network was devised. An empirical research of this model was done, based on the date from 10 projects in cooking industry, in 2000. The results showed the wavelet network training up to 7933 times with convergence precision of 0.001, taking 3.31 seconds.

Proper key renewal interval may control the trade-off between security and renewal cost, and thus partial and gradual leakage of key information is studied for predicting quantitatively key life. A key is treated as a random vector. Smooth entropy represents still secret almost uniform random bits that can be extracted with high probability from a compromised key (as a result, smooth entropy may be used to define the key life). The varying of smooth entropy over time is modeled as a compound Poisson process.

The drawbacks of traditional web information search mode is analyzed and summarized in this paper. On this basis, the mobile agent technology is further analyzed and studied. Accordingly, a kind of new-type web information search mode utilizing mobile agent technology is constructed. Applying mobile agent technology to realize the web information search system can save much network bandwidth, and when mobile agent runs on the server, it doesn't need the persistent connection with client server, so it has much flexibility and reliability.

Active network is a new framework where network nodes not only forward packets, but also perform customized computation on the packet flowing through them. It provides a programmable interface to the user where users dynamically inject services into the intermediate nodes. However, the traditional prototype of network management does not accommodate to the management of active networks, it cannot utilize the distributed computing capabilities that active networks provides. This paper analyses the structure and mechanism of the active network management system, introduces a pattern of active network management, and studies the structure, management mechanism and studies the structure, management mechanism, design outline and each connection of the management system. The paper also studies the network topology discovery and traffic.

Flight security is directly influenced by flight plan conflicts. In this paper, various factors which cause conflicts in flight plan are discussed. Then, an optimization math model and the algorithm for detecting these conflicts are established. This algorithm is proved to be highly efficient in practice.

Neural networks have good learning ability, and wavelets have good time-frequency localization properties. Combining the advantages of neural network and wavelet, wavelet neural network (WNN) shows great advantages than general neural network. WNN has been successfully applied in many areas. But in Intrusion Detection System (IDS), how to apply WNN is still one question. In this paper, we give a WNN-IDS model that using WNN instead of general neural network. Results show that applying WNN to IDS can shorten the training time and increase the test accuracy.

Introduces some refer conceptions of the PCI bus and issues the finite machine design in PCI bus interface controller. Implements a PCI bus controller in programmable device FPGA by using VHDL language. The result of test indicates that this design accords with sequence request of PCI criterion and get a good effect.

There are a series of advantages about the double involute gear with ladder shape of tooth. But whether the ladder-shaped of the double involute gear will cause the increase of the vibration or bring some other vibration components? This problem is especially concerned by people all the while. The essence of wavelet packet analysis is to make further decomposition of wavelet decomposed result, so the analysis will yield much better frequency localization. Duo to wavelet packet has agile and changeful characteristics in disposing signals, it can be used to analyzed and compared the vibration characteristics of common involute gear and double involute gear conveniently and exactly by decomposing and reconstructing signal components of correlative frequency components. The result shows that the double involute gear has lower noise, smaller vibration and better dynamics characteristic than common inlovute gear

Wireless sensor network has broad applications in target tracking and locating, especially fit for military detection or guard. By establishing a tracking cluster, this article proposes a Tracking Cluster Rekeying Protocol (TCRP). Sensors can locate the moving object in the monitored area based on given sensitivity and form a tracking cluster around it. This tracking cluster can follow the target logically, process detected data and report to the sink node, and thus achieve the tracking function. To improve its security and applicability, cluster session key is used for data exchange. And pre-established encrypt links are used to guarantee the whole system's security.

Cryptographic key is the central problem in cryptographic techniques. The key based on password protection is not enough secure because of the low entropy in user chosen passwords that can be exploited to launch password-guessing attacks. And the length of cryptographic key generated from user's biometric features directly is limited. Differ from prior methods, a novel scheme of cryptographic key generation based on biometric features and secret sharing is proposed. Biometric feature vectors is transformed through orthogonal matrix, and mapped to integer spaces. Then, a steady integer vector is obtained, and can be utilized to bind with cryptographic key by Shamir's secret sharing scheme. This method may realize the security of key storing, and can produce the key of the arbitrary length.

This paper presents a new kind of Real-time Control Protocol to solve the problem of nondeterministic network traffics on Ethernet, expatiate to the basic theory of the protocol, prove that this protocol can help the Ethernet implement transmit real-time data by experiment and implement the protocol Stack in RTLinux OS

Owing to a thorough survey and comprehension of the education of E-commerce courses in China, U.S. and Britain, the authors provide some ideas about the E-commerce teaching reform including: course plan, structure and content, teaching methods, inter-discipline, teachers. This article illustrates the direction of the E-commerce course reform—the combination of science and engineering, technology and management, natural science and human arts, enterprises and education units. The essence is the direct incorporation of theory and practice under the instruction from professors of economic management and engineering technology.

Traffic models play a significant role in the analysis and characterization of network traffic and network performance. Thorough research and accurate modeling on network traffic become an efficient way to explore network internal mechanism and optimize network performance. In this paper, we put forward an Haar DWT-based (Discrete Wavelet Transform) traffic model. The properties of the DWT-based model and the scaling analysis on the two simulated traces show that our model can capture the self-similar and multifractal properties in the network traffic.

As we know, a challenge of image denoising is how to preserve the edges of an image when reducing noise. In this paper, by showing the model of noisy images and taking advantage of the multiresolution analysis with wavelet transform to remove the noise, we propose a wavelet image thresholding scheme. The size of the threshold is interrelated with the noise degree, and then we present an efficient denoising method. Experimental results demonstrated that this algorithm could achieve both good visual quality and high SNR for the denoised images.

In this paper, a novel oil pipeline leakage detection system has been presented. A modified multi-scale wavelet transformation modulus maxima algorithm was used to detect and locate the singularity of the characteristic signal of oil pipeline. According to the singularity, the leakage point could be known. In order to promote the precision, an adaptive threshold wavelet packet algorithm be utilized to filter the noises, after denoising the essential singularity of signal would not loss. Through simulation experiments confirm the validity of these algorithms.

In the process of data embedding and extraction,there are a little articles that change the host media too much.In this paper,we present a method for still image steganography in which we change the host image into Gaussian white noise for increase the quantity of embedden information.In the process of extraction,we use the method of denoise, regard the host image as noise,take advantage of the character of Gassian white noise to delete the host image for getting the hidden image.Method of denoise used for extraction is a new view in steganography.

CSCL experiment platform shall provide the collaborative learning activities and/or experiments with the necessary support, which obviously requires the platform to own remarkable flexibility. Mainly centering on the resources and/or tools, the existing CSCL system can only fit the collaborative learning activities in specific fields, but not those in other fields. The practices from the CSCL Experiments direct our attention to that an excellent support platform of collaborative learning shall be “flexible”. This paper focuses on how to build a “flexible” CSCL experiment platform, discussing the connotation of the “flexibility” and lunching a train of ideas and methods. CSCLEP, a project developed by adoption of these ideas, is introduced as well.

The key point for shape-based image retrieval is to detect and extract the object edge, i.e., get the object shape or outline of the image. In terms of the shape features, the feature vector of the shape can be generated, then, the similarity matching is carried out among the feature vectors. This paper presents the wavelet algorithm to extract the edge of the object from the image. Since each pixel of edge is of orientation, the orientation angle of pixel can be calculated in the range of 0∼2π and generate the histogram to describe the shape. As the feature vector, the histogram is used to the image matching retrieval.

Due to the limited bandwidth of wireless transmission, in this paper we present an efficient scheme of wireless monitor system. The scheme adopts an improved fast SPIHT algorithm to encode video image in combination with the strategy of ROI (Region of interest) with priority. With regard to the problem how to separate ROI and non-ROI, we introduced a background modeling method based on statistics, which can adapt to the changing environment. Furthermore, the scheme produces an embedded stream, supports successive transmission, the rate of coding image can be accurately controlled; it not only rationally utilized network bandwidth, but also made the best of wavelet's advanced speciality.

Item-selecting method of maximum information is widely used in CAT. But it inclines to select the item of high differentiation, hence brings about high percentage of exposure and low percentage for using item base. A-stratification method can overcome this, but it has its faults. The paper introduces two modified item-selecting methods based on a-stratification item-selecting. On the basis of analyzing the two kinds of methods, the paper compares them by monte-karlo simulation experiment.

Non-destructive testing is developing in the direction of automatization and intelligentization, but there still exists some difficulty in analyzing quantitatively the signal of magnetic flux leakage. The Mallat quick algorithm of wavelet transformation is described simply. As an advanced digital signal processing method, wavelet transformation can be used to compress data and separate the signal and the noise in the magnetic flux leakage inspection. So introducing wavelet transformation into the magnetic flux leakage inspection provides high feasibility for the quantitative analysis of the test signals.

Traditional intrusion detection methods lack extensibility and adaptability in face of unknown attack types. At the same time, current neural networks algorithms are generally based on supervised learning, and need labeled data for training first. Self-organizing map algorithm is used to detect novel attack in some papers[2,4,5].In this paper, self-organization map is adopted to classify the input data produced by detectors, followed by a updated learning vector quantization–LVQ3. Experiment results indicate that this algorithm is more effective for intrusion detection than traditional intrusion detections or the algorithm that only utilizes SOM.

In establishing Gas-storage in the water saturation oil-field, the composition of gas produced and injected varies as driving of injected gas in each layer. Adopting the oil-gas Chromatogram Fingerprint Technology can monitor the gases of gas storage-layer and production preferably. The analyzing of the oil-gas fingerprint characteristics of the different layer makes sure the fingerprint differences of these oil-gas and reasons of bringing the differences, and develops the prescription experiments of oil-gas of two layers and three layers, establishes the mathematic models of the each layer's productivity to compute, and applies wavelet pack analyzing technique to increase the oil-gas fingerprint calculation's accuracy. The computing result indicates that the error margins of the actual prescription and the fingerprint calculation are within 6% for two layers or three layers joined injection and extraction. It shows that the oil-gas fingerprint can be used to calculate the productivity contributions of each layer in the oil-well with many layers joined to inject gas and produce, and monitor the productivity's variety of each layer to storage gas.

The prediction of the yield of gas storage has been taking the important position in gas storage program decision. The research of prediction method also has never stopped. Though the yield of gas storage experiences waved process, which is from rising to steady and to descend, for the nature geology of gas storage is different, the explore method is different, the explore measure that adopted is different, the yield of gas storage present very complicated variability. This paper applies the rescaled-range(R/S) analysis method that is put forward by Hurst to predict and analysis the time sequence of yield of gas storage.

With the development of the technology of information security, iris recognition based on biometrics has become more and more important. In an iris recognition system, preprocessing, especially iris localization plays a very important role. So far, there are many iris localization algorithms having been proposed. In this paper, we propose an iris localization algorithm, in which we localize iris through detecting the edge points and improved integral differential operator and curve fitting. All the procedures of the algorithm are proved to be valid through our experiment on 648 iris images from CASIA (The Institute of Automation, Chinese Academy of Sciences) database.

Reviewing the history of selection of machining parameters for the metal cutting operations, based on the principle of Artificial Intelligence, an optimal selecting system is an brought forth which uses a complex method to search for the resolution of metal cutting parameters involved in the nonlinear machining conditions and constraints. The target function comprises two aspects: the minimum time for machining a workpiece and the maximum profit rate. Detail of the hierarchy and structure of the system is presented with the direct-depth inference strategy. It has been proved by a series of trials of workshop processes that this system is workable and able to provide machining engagements at the expert level.

The relation between the differential equations in a special abstract function space and the integral equations in the space is discussed by using the continuous wavelet transform; the result that the differential equations can be transformed into the integral equations is obtained; they are equivalent not only in the weak topology but also in the strong topology.

Introduces the pins, structure and function of flash memory: K9F1208UOM. The data storing part of an embedded system is designed by this flash memory and its basic operating principle and interface technique is researched in this paper; it also gives the interface circuit of hardware and realization of software. This paper provides a scheme with best cost-benefit for large data storage market, and it has wide applied foreground in field of storage application.

The during the gear fault detection and diagnosis, the fault feature extraction is the key of diagnosis, but it is multifarious for the method of fault feature extraction. On the basis of continuous wavelet transform, this paper mainly introduces that the basic principle of wavelet transform and its application in feature extraction for fault diagnosis are set forth. It is aimed at no steady state pulse signal feature when transmission gear have fault, select suitable wavelet, apply wavelet transformation successfully to diagnose crackle fault of transmission gear, which indicates the advantage of wavelet analysis. The results show that the continuous wavelet transform is a useful tool in diagnosis of crack in gears. The new method for prognosis and diagnosis of incipient small crack can be applied to engineering practice effectively.

The signal gathered often contains the noise in the actual project, so only can extract the useful information in the original signal by doing noise reduction. Wavlet method have low entropy, multi-resolution characteristic, not pertinence, choose base function flexibly, adaptive time-frequency localization characteristic. All these characteristics that Fourier have not, So Wavlet analyze can extract useful information from the original signal by reducing noise.

The signal gathered in the actual project belonged to two classes, steady signal and non-steady signal.

Author Index