Applied Mechanics and Materials Vols. 385-386

Abstract: With the development of WSN (Wireless Sensor Network) technology, WSN node requires its volume to become gradually small. Antenna of WSN node occupy the most area, therefore this paper puts forward minimization design for antenna of WSN node. A printed dipole antenna is designed and then by using antenna flexural method, the antenna structure is improved with minimization design. Under the condition of several performance indexes of the antenna can be met, the antenna area after minimization is decreased by about 39%.

Abstract: As a kind of UNB modulation method, time-phase modulation has a specialty that its power spectrum is concentrated around the carrier frequency. Thus, it can work within a narrow channel interval without interfering the conventional communication system. In this paper, the performance of multiple users was analyzed for time-phase modulation based on the correlation coefficient. It can be found that wider carrier frequency difference between adjacent users is necessary for time-phase modulation due to the modulated information consisted in the very wide sideband. Moreover, the influence between multi-users will fluctuate versus the difference of carrier frequency, which diminishes as a whole when the difference of carrier frequency increases.

Abstract: t is known that the performance of a signal constellation used to transmit digital information over the additive white Gaussian noise channel can be improved by increasing the dimensionality of the signal set used for transmission. This paper derives an algorithm for constructing codes for six-dimensional signals over finite fields of the algebraic integer ring of the cyclotomic field modulo irreducible elements with the norm , where is a prime number and or .These linear codes can correct some types of errors and provide an algebraic approach in an area which is currently mainly dominated by nonalgebraic convolutional codes.

Abstract: Sequential pattern is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining sequential patterns focus on the frequent sequences, which do not consider the infrequent sequences and lifespan of each sequence. On the one hand, some infrequent patterns can provide very useful insight view into the data set, on the other hand, without taking lifespan of each sequence into account, not only some discovered patterns may be invalid, but also some useful patterns may not be discovered. So, we extend the sequential patterns to the indirect temporal sequential patterns, and put forward an algorithm to discover indirect temporal sequential patterns in this paper.

Abstract: State estimation of power system is the basis of all the high level application and analysis for dispatch centers. Against the bad convergence of the WLAV (weighted least absolute value)-based state estimation method, a new robust state estimation method based on exponential weight function (E-LAV) is presented in the paper. This method uses an exponential weight function to replace discontinues weight function of WLAV to improve the poor convergence. The results of the sample system of 4-node system and the IEEE 118-node system show that the E-LAV-based state estimation method not only owns the similar convergence as WLS-based method, and also owns a strong robustness. Through modifying the weight matrix of the least squares, the proposed new method has good prospects for application in practical engineering.

Abstract: The presence of aircraft voice alarm system voice recording work is limited by the staff, environmental and complicated operation, not applying to automatic test, difficult secondary development and poor versatility, for which we set up a library of standard voice using voice IC, using computers to control the voice of the call need information, combined with speech synthesis technology, which can synthesize voice alarm information corresponding to the full with the desired model aircraft. This paper elaborates on the basic principles of the system, the establishment of a process and the realization of human-computer interaction. The device is very easy to communicate with other devices, if used for a different system, without changing the hardware circuit, simply by changing software and voice can be required, good portability and versatility.

Abstract: To address the problem of data non-synchronization to multi-radar, a new time registration approach is presented in this paper, in which the data is synchronized using Kalman based Newton interpolation (KNI). In this work, a preliminary alignment is firstly implemented with Newton interpolation, providing a dummy measurement at each time step. Secondly, the Kalman filter is followed, in which the measurement innovation is generated by the dummy measurement. And within the framework of Kalman filter (KF), the data is synchronized into the datum step. Thus, using the most of recent measurements, this new proposed method can realize the synchronization of multi-radar data and as a result, improve the performance of data fusion and target association. An application example is given to draw a comparison between this new algorithm and the existing algorithm. The simulation results show the efficiency of this algorithm.

Abstract: Speech segregation based on energy has a good performance on dual-microphone electronic speech signal processing. The implication of the binary mask to an auditory mixture has been shown to yield substantial improvements in signal-to-noise-ratio (SNR) and intelligibility. To evaluate the performance of a binary mask based dual microphone speech enhancement algorithm, various spatial noise sources and reverberation test conditions are used. Two compare dual microphone systems based on energy difference and machine learning are used at the same time. Result with SNR and speech intelligibility show that more robust performance can be achieved than the two compare systems.

Abstract: In this paper we introduced the application of Fuzzy KDA in speech emotion recognition using elicited data. The emotional data induced in a psychology experiment. The acted data is not suitable for developing real world applications and by using more naturalistic data we may build more reliable system. The emotional feature set is then constructed for modeling and recognition. A total of 372 low level acoustic features are used and kernel discriminant analysis is used for emotion recognition. The experimental results show a promising recognition rate.

Abstract: Extracting voltage flicker from the sampling voltage signal is a precondition for management of flicker. Voltage flicker signal is a low frequency time-varying non-stationary signal. The traditional fourier transform has great limitations when analyze the non-stationary signal for not having the time resolution. As wavelet transform has good property of time-frequency localization, it become a powerful tool for analyze this kind of signal. This paper adopts multi-resolution analysis of wavelet to extract voltage flicker signal. Furthermore, according to the characteristics of wavelet function, this paper selects Daubechies wavelet to accomplish the multi-level decomposition and reconstruction of signal, in order to get the frequency and amplitude of voltage flicker signals. Based on the principle of modulus maximum, it can be determined what time the voltage flicker happen and over. The results of MATLAB simulation indicate that voltage flicker signal can be effectively extracted by wavelet multi-resolution analysis. Wavelet multi-resolution analysis is considerably ideal for voltage flicker extraction.