Applied Mechanics and Materials Vols. 719-720

Abstract: Hyperspectral remote sensing technology provides a new way to identify red tides types, but many existing methods can’t take full advantage of the spectral reflectance characteristics and often yield false recognitions. So, on the premise of perfect spectral curves library of red tides to be referred, this paper proposes an algorithm based on spectral reflectance characteristics and wavelet decomposition for red tides recognition. The algorithm identify the red tide species by applying wavelet analysis to a certain wavelength range limited by the spectral features. To compare and prove the effect of this algorithm, do simulate experiments with both the proposed method and the traditional SAM method. The results show that, compared with SAM method, the algorithm put forward in this paper can better indentify the species of red tides.

Abstract: Sensorineural hearing impaired individuals have narrow hearing range, especially in high frequency. Due to acoustic diffusion masking, just amplifying speech can’t meet patients’ demands. Traditional methods restore the signal of high frequency in low frequency, which has two weaknesses. On the one hand high frequency covers low frequency, on the other hand the methods change the energy of speech. These two aspects severely decrease speech loudness, perception and recognition. To solve the above problems, we have proposed an improved frequency shifting compression method based on Auto Energy Gain Compensation (AEGC) which is used for hearing aids. Fully investigating the relationship of speech frequency with energy and recognition, the scheme selectively keeps the information of vital frequency band. Meanwhile, it compensates the speech loudness with AEGC module. With the proposed scheme, it has obviously improved the speech perception and recognition. Both the theoretical simulation result and subjective test in APP on android platform show the energy is compensated and the speech recognition ratios are enhanced 20%-30% in different environments, especially efficient in quiet, noisy and normal environment.

Abstract: Recently, there is a growing trend to investigate wide dynamic range compression technique with the development of hearing aids. However, existing methods used to enhance expected speech as well as the background noise and classical linear compression can’t achieve the requirements of hearing impaired individuals’ reaction to the loudness compensation flexibly. To cope with the problems, this paper proposes a new approach to suppress the enhanced background noise and to dismiss abnormal sound caused by severe jittering from high frequency signal. What’s more, a scheme of curve compression is introduced to improve the loudness amplifying flexibility instead of conventional linear compression. Both the theoretical simulation and testing on cellphone hearing aid APP based on android system prove the proposed method can improve the performance of speech enhancement and provide more choices to strength sound with the curve compression ratio.

Abstract: Speech processing in hearing aid often operates in noisy environments. So noise reduction algorithm is very important for hearing aid. This paper reviews conventional wiener filter algorithm and points out its remained problems----musical residual noise and low speech intelligibility for low input signal-to-noise ratios (SNR). To solve the problems, the proposed algorithm divides the frequency (0-8000 Hz) into 16 bands in Mel-frequency scale and processes the signal separately to remove musical residual noise. Meanwhile, a new Voice Activity Detector (VAD) applied for digital hearing aid is also presented. Based on the accurate judgment of speech frame, the filter processes the speech frame and non-speech frame in different ways. The method is realized in matlab to test output SNR and PESQ (Perceptual Evaluation of Speech Quality). It is also realized in android app for real-time test. Both of the experimental results demonstrate the advances of the proposed algorithm.

Abstract: The tight rock often has low porosity, low permeability and poor pore connectivity, which it is difficult for formation evaluation. Nuclear Magnetic Resonance (NMR) logging is widely used in fluid typing and reservoir parameters determination to provide the information of porosity, permeability and pore size distribution. NMR relaxation mechanisms are characterized by the pore-scale petrophysical models. Monte Carlo algorithm describes the Brownian motion of fluid molecules in pore space. In the paper we setup a 3D digital core of the tight sandstone with X-ray computer tomography (CT) images to model NMR response with Monte Carlo random walk algorithm. We compared T₂ distributions from the numerical calculated pulse echo trains from the measurements. The results show that the simulated NMR response is consisted with the experiment.

Abstract: A revise system on frequency response based on power amplifier of shortwave transmitter is designed to solve the problem of nonlinear response at different frequencies on power amplifier. This system is able to guarantee the output of power amplifier to fluctuate slightly around the rated power though the two processes named pre-revise on frequency response of power amplifier and current revise on frequency response of power amplifier. A modified variable step and multi-rate updated LMS algorithm is put into use to reduce the calculated amount and boost the rate of the revise. It is demonstrated by experiment that the revise system of this paper is able to guarantee the output of power amplifier around the rated power with the deviation of dB at a high convergence rate.

Abstract: Differential SAR interferometry (DInSAR) provides a new tool for detecting deformation in the line of sight (LOS) of Earth’s surface. However, not all deformation can be detected. In order to get the maximum detectable deformation, the decorrelation in terms of coherence and deformation has been proposed. The influence of deformation on decorrelation is then discussed for both registering drifted pixels method and registering fixed method. Based on aforementioned analysis, the limitation of maximum detectable deformation has been derived for these two cases. Furthermore, computer simulations are carried out to confirm the detectable deformation discussion as the coherence equals to 0.5. The result of registering drifted pixels method shows that the detectable limitation does not only depend on coherence but also on other factors such as beamwidth. The result of registering fixed pixels method, however, shows that the detectable deformation and baseline have an inverse relationship.

Abstract: An important element of military operations is the evaluation of courses of action (COAs) in order to overcome emergency situations. In this kind of decision-making situation, it is important to consider several criteria to evaluate the various COAs, and to compare and rank the COAs according to their ability to meet the selected criteria. It is often necessary to balance the importance of these criteria accordingly to commander’s viewpoints. A new method has been proposed for evaluating the COAs that is based on the possibility calculus known as stochastic interval weighted average operator. The method first presents the attributes and its weights by stochastic interval numbers, so empirical and statistical information, as well as for opinions and judgments made by commanders and analysts, all can be integrated, second, the random numbers of sums of independent variables can be generated by Monte Carlo method, lastly, the comparing and sorting of COAs are realized by random variable Comparison. An example is given to demonstrate the feasibility and practicability of the proposed method.

Abstract: It’s important to get accurate carrier phase and frequency information when using a standalone GPS receiver. In weak signal applications, to keep a stable tracking is hard to achieve because measuring error will be huge when the SNR is low. Different methods are used to improve the SNR before the detector in a tracking process, such as coherence integration. And this paper keeps eyes on a different viewpoint, on how to refine estimation results. A cascade structure is introduced for weak signal tracking. This structure is divided into two levels. In the first level, raw phase estimation and accurate frequency estimation is provided to achieve stable work in low CNR environment. In the second level, the raw phase estimation is refined to achieve accurate tracking requirement. This cascade structure can also work jointly with any other SNR-improving technology to get a better performance.

Abstract: Airport runway Foreign Object Debris (FOD) jeopardizes flight safety and leads to a large amount of financial cost on flight maintenance constantly. Several FOD detection systems based on a variety of detection techniques and architectures have been developed. This paper gives a brief introduction of our FOD detection system, in comparison with FOD detection systems currently in the international market. What distinguish our system from all the others is that, our detection approach is based on both linear frequency modulated continuous wave (LFMCW) radar signal and image data. This innovation combines the advantages of radar in shape detection and image in appearance detection. As a result, it increases the FOD detection rate and reduces the false alarming rate. Experiments following Federal Aviation Administration (FAA) advisory indicate that our system has reached the FAA requirements.