Applied Mechanics and Materials Vols. 333-335

Abstract: Scheme for the implementation of wavelet filter using switched-current (SI) circuits is presented in this paper. The circuits are composed of analog filter bank whose impulse responses are the first derivative of Gaussian and its dilations. The transfer functions of the basic wavelet filter can be given by Pade approximation. Therefore wavelet filter can be conveniently implemented by SI circuits. Simulated results indicated a good performance of wavelet filter.

Abstract: Laser altimeters are used to measure the elevation and detect the topography of terrain, which have been widely applied in military and civilian field. In the course of measurement, errors will inevitably generate as a result of ground settlement, surface relief and tilt, etc. In order to reduce measurement errors, the technology of laser waveform digitalization was researched. Firstly, the principle of error generation was introduced. Secondly, the basic scheme of signal processing system for digital laser altimeter was present. System components and signal processing procedure were researched. Next, the technology of extracting echo waveform parameters was discussed. Finally the method of using digital waveform parameters to correct measurement errors was provided. The principle of target classification and identification based on echo waveform parameters was introduced also.

Abstract: Interfering noise of power line is one of the important factors which affects the quality of power line communication (PLC). Its frequency spectrum has the character of the 1/f process and the great autocorrelation. The wavelet analysis is an important signal-processing tool. Selecting suitable wavelet analysis can turn non-white noise to white noise, followed by wiener filtering, we can achieve the purpose of denoising. This paper introduces a denoising method of combining wavelet analysis with wiener filtering. Experiment proves this method has a strong feasibility and practical value.

Abstract: Compressive Sensing (CS) Theory enables sampling discrete signals with quite lower sampling rate compared with traditional Nyquist sampling rate and guaranteeing faithful reconstruction. Based on CS theory, Analog-to-Information Conversion (AIC) was proposed to process continuous-time signal. In this paper, the framework of Analog-to-Information Converter is composed by a pseudo-random demodulator, a low pass analog filter and a low speed sampler. And we mainly discuss the damage on the signal recovery produced by lower and higher orders of filter impulse response.

Abstract: Interference monitoring and analysis for GNSS frequency bands plays an important role in construction and development of satellite navigation systems, which can promote interference source locating, and has much benefit for system construction and the development of anti-jamming equipments. Due to high satellite orbits, GNSS signals reached the ground are very weak and submerged below the thermal noise, which makes it vulnerable to interference. Interference sources for satellite navigation system require only a small transmission power; however, a significant interference effect can be obtained. Therefore, a high sensitivity is needed by interference monitoring for satellite navigation system. The interference judgment threshold is close to thermal noise power, which often causes a higher probability of false alarm. It is very important to reduce the probability of false alarm at the same time to ensure high sensitivity. In this paper, a high sensitivity (low missed detection probability) and low false alarm interference monitoring method is proposed, a dual decision threshold is designed, thus the probability of false alarm can be effectively reduced at the same time of identifying interference accurately. The experimental results demonstrated the effectiveness of the algorithm.

Abstract: To solve the effect of the range imaging broadening, the speed estimation of high-speed non-cooperative target of inverse synthetic aperture lidar translated for the estimation of the frequency modulation slope. A parameter estimation method of fractional Fourier transform (FRFT) was presented. This method could effectively estimate the frequency modulation slope. Under low SNR, the error of the estimate was less than 0.03. The research laid the foundation for velocity compensation and 2-D range-Doppler imaging of the follow-up.

Abstract: To improve anti-ship missiles selection capability over the intended target of the ship formation, a method is presented to select intended target by utilizing the mutual locations and RCS information of the ship formation. Firstly, this method takes advantage of the hausdorff distance to match the optimum position of the point set which are respectively acquired by fire-control radar and terminal guidance radar. Then, the point sets common points are acquired by the bidirectional nearest neighbor rule. At last, the common points are further selected by utilizing the RCS information of the target. The composite metric distance method is presented by utilizing the targets RCS information and the formation information, and the selection probability of the intended target is analyzed. The simulation experiments show that this method is free from the influence of the dispersion error of the control end and the whole maneuver of the formation target, and its correct selection probability over the intended target is higher than the method which uses the formation information only.

Abstract: Orthogonal discrete frequency coding waveforms(DFCWs) is an ideal quasi-orthogonal waveform. This paper analyses the ambiguity function of DFCWs and the results indicate that the correlation characteristics of DFCWs have only relationship with the code length and coding order. Based on the theoretical analysis, we defined the cross-correlation energy as the cost function, and used the improved discrete particle swarmoptimization(DPSO) to optimize the order of DFCWs. The new signal can effectively restrain the cross-correlation level between the two DFCWs. Simulation results verify the effectiveness of the designed DFCWs.

Abstract: Due to the gear fault signal are often covered by complex noisy even under varying-speed running condition, and it is difficult to simulate the gear fault impact in practical work. To address the problem, a simulation scheme of gear fault is proposed in this paper. In the proposed approach, the gear fault model is built in UG at first. Subsequently, the model is imported in ADAMS, in which the meshing force instead of the vibration signal from a gearbox in real world is extracted, and then the envelope of the meshing force signal is extracted. Finally, the envelope order spectrum analysis scheme is used to obtain the feature order of gear faults. Simulation is conducted to verify the validity of the proposed method.

Abstract: Wind turbine transmission system with complex structure and abundant fault features, also the fault features are variable, three typical fault and failure forms of rolling bearings, shafts and gears in wind turbine drive train were analyzed, also the failure mechanism and corresponding vibration signal characteristics was proposed. In the wind turbines transmission system, the vibration signal can reflect most of the fault information, as there was non-stationary signals in the vibration signals and wavelet transform for feature extraction was proposed. The whole framework of the fault diagnosis platform was constructed, the lower machine based on the measurement unit and the PC based host computer diagnosis system was designed, the diagnosis system was powerful and easily cut, which was strong help to the fault character extraction, also it was practical and available for wind turbine fault diagnosis.