Zero-crossing Rate Research Articles

Spectral Zero-Crossings: Localization Properties and Applications

We present an analysis of the rate of sign changes in the discrete Fourier spectrum of a sequence. The sign changes of either the real or imaginary parts of the spectrum are considered, and the rate of sign changes is termed as the spectral zero-crossing rate (SZCR). We show that SZCR carries information pertaining to the locations of transients within the temporal observation window. We show duality with temporal zero-crossing rate analysis by expressing the spectrum of a signal as a sum of sinusoids with random phases. This extension leads to spectral-domain iterative filtering approaches to stabilize the spectral zero-crossing rate and to improve upon the location estimates. The localization properties are compared with group-delay-based localization metrics in a stylized signal setting well-known in speech processing literature. We show applications to epoch estimation in voiced speech signals using the SZCR on the integrated linear prediction residue. The performance of the SZCR-based epoch localization technique is competitive with the state-of-the-art epoch estimation techniques that are based on average pitch period.

A Preliminary Study on the Feature Distribution of Deceptive Speech Signals

A preliminary study is conducted to compare the feature distribution between normal and deceptive speech, and the results are reported in this paper. The objective of this research is to show that deceptive speech may be recognized through the acoustic parameters of general speech characteristics. Six speech parameters, i.e., Mel-frequency Cepstral Coefficients (MFCC), Relative Spectral Filter Perceptual Linear Prediction (RASTA-PLP), pitch frequency, time-domain samples, zero-crossing rate and fractal dimension are used in the statistics. The distributions of these parameters indicate clear differences between the two speech styles. The lowest average degree of difference for these features was 4.74%, and the highest degree was over 20%. Therefore, the distribution demonstrates that there is significant distinction between speech relating the truth and speech relating falsehoods. Linear Discriminant Analysis (LDA) and the Gaussian Mixture Model (GMM) are used to recognize the two psychological states of people’s pronunciation, with accuracy above 50%. The results show that there is in fact deceptive information in speech signals and that it can be detected by pattern recognition. These findings provide the theoretical basis for detecting deception in speech signals.

High-Speed Spindle Fault Diagnosis with the Empirical Mode Decomposition and Multiscale Entropy Method

The root mean square (RMS) value of a vibration signal is an important indicator used to represent the amplitude of vibrations in evaluating the quality of high-speed spindles. However, RMS is unable to detect a number of common fault characteristics that occur prior to bearing failure. Extending the operational life and quality of spindles requires reliable fault diagnosis techniques for the analysis of vibration signals from three axes. This study used empirical mode decomposition to decompose signals into intrinsic mode functions containing a zero-crossing rate and energy to represent the characteristics of rotating elements. The MSE curve was then used to identify a number of characteristic defects. The purpose of this research was to obtain vibration signals along three axes with the aim of extending the operational life of devices included in the product line of an actual spindle manufacturing company.

A coder with adaptive filters and non-uniform techniques for simple speech communication

This paper proposes a new method of non-uniform sampling coding that utilizes a modified zero-crossing rate filter and a variable filter. In speech signals, the modified zero-crossing rate refers to the number of times in which the signal crosses zero in a single frame; the rate is high in noisy sections but low in sections of vocal sounds that are quasi-periodic. And the modified zero-crossing rates of unvoiced signals are similar to those of voiced signals. In speech signals, the non-uniform sampling coding method utilizes the peaks and valleys in a frame. In noisy sections, there are a lot of zero crossings and thus creating many peaks and valleys, while sections with voice sounds have zero-crossings numbering only 2–3 times of their fundamental frequency. The method proposed in this paper utilizes the number of peaks and valleys in determining the validity of a sample; the less we have peaks and valleys, the higher compression rates we get. Therefore, by decreasing the number of samples from sections with noise and voiceless sounds which have a high number of peaks and valleys, and by creating a signal that has little effect on the recognized signals, we may acquire a better compression rate. As examined and proposed in this paper, these features can be used to design a filter coefficient for a better performance in order to improve compression rates of non-uniform sampling coding and to maintain high quality signals.

The effect of CPAP treatment on EEG of OSAS patients.

Continuous positive airway pressure (CPAP) is currently the most effective treatment method for obstructive sleep apnea syndrome (OSAS). The purpose of this study was to compare the sleep electroencephalogram (EEG) changes before and after the application of CPAP to OSAS patients. A retrospective study was conducted and 45 sequential patients who received both polysomnography (PSG) and CPAP titration were included. The raw data of sleep EEG were extracted and analyzed by engineers using two main factors: fractal dimension (FD) and the zero-crossing rate of detrended FD (zDFD). FD was an effective indicator reflecting the EEG complexity and zDFD was useful to reflect the variability of the EEG complexity. The FD and zDFD indexes of sleep EEG of 45 OSAS patients before and after CPAP titration were analyzed. The age of 45 OSAS patients was 52.7 ± 5.6years old and the patients include 12 females and 33 males. After CPAP treatment, FD of EEG in non-rapid eye movement (NREM) sleep decreased significantly (P < 0.05), while FD of EEG increased in rapid eye movement (REM) sleep (P < 0.05). Meanwhile, zDFD were decreased remarkably in both NREM and REM sleep after CPAP therapy (P < 0.05, respectively). CPAP therapy had a significant influence on sleep EEG in patients with OSAHS, which lead to a more stable EEG pattern. This may be one of the mechanisms that CPAP could improve sleep quality and brain function of OSAS patients.

Classification of Vehicles Based on Audio Signals using Quadratic Discriminant Analysis and High Energy Feature Vectors

The focus of this paper is on classification of different vehicles using sound emanated from the vehicles. In this paper,quadratic discriminant analysis classifies audio signals of passing vehicles to bus, car, motor, and truck categories based on features such as short time energy, average zero cross rate, and pitch frequency of periodic segments of signals. Simulation results show that just by considering high energy feature vectors, better classification accuracy can be achieved due to the correspondence of low energy regions with noises of the background. To separate these elements, short time energy and average zero cross rate are used simultaneously.In our method,we have used a few features which are easy to be calculated in time domain and enable practical implementation of efficient classifier. Although, the computation complexity is low, the classification accuracy is comparable with other classification methods based on long feature vectors reported in literature for this problem.

An Improved Positioning Algorithm With High Precision for Dual Mach–Zehnder Interferometry Disturbance Sensing System

An improved positioning algorithm for dual Mach–Zehnder interferometry (DMZI) disturbance sensing system is proposed. We employ zero-crossing method, which can be computed easily to extract the disturbance signal segment with maximum zero-crossing rate. Meanwhile, we use general cross correlation based on Wiener filtering and $G_{\rm nn}$ subtraction weighting function (WG-GCC) to estimate the time delay of the extracted signal, which is robust to the correlated noise. Finally, we experimentally demonstrate that the proposed positioning algorithm can greatly improve the positioning accuracy with positioning error of ±20 m. Compared with the traditional positioning algorithm, the positioning error has been reduced by an order of magnitude. This algorithm has a promising potential in real-time fence perimeter applications.

Broad Phoneme Classification Using Signal Based Features

Speech is the most efficient and popular means of human communication Speech is produced as a sequence of phonemes. Phoneme recognition is the first step performed by automatic speech recognition system. The state-of-the-art recognizers use mel-frequency cepstral coefficients (MFCC) features derived through short time analysis, for which the recognition accuracy is limited. Instead of this, here broad phoneme classification is achieved using features derived directly from the speech at the signal level itself. Broad phoneme classes include vowels, nasals, fricatives, stops, approximants and silence. The features identified useful for broad phoneme classification are voiced/unvoiced decision, zero crossing rate (ZCR), short time energy, most dominant frequency, energy in most dominant frequency, spectral flatness measure and first three formants. Features derived from short time frames of training speech are used to train a multilayer feedforward neural network based classifier with manually marked class label as output and classification accuracy is then tested. Later this broad phoneme classifier is used for broad syllable structure prediction which is useful for applications such as automatic speech recognition and automatic language identification.

A Contraction Mapping Approach for Robust Estimation of Lagged Autocorrelation

We consider the zero-crossing rate (ZCR) of a Gaussian process and establish a property relating the lagged ZCR (LZCR) to the corresponding normalized autocorrelation function. This is a generalization of Kedem’s result for the lag-one case. For the specific case of a sinusoid in white Gaussian noise, we use the higher-order property between lagged ZCR and higher-lag autocorrelation to develop an iterative higher-order autoregressive filtering scheme, which stabilizes the ZCR and consequently provide robust estimates of the lagged autocorrelation. Simulation results show that the autocorrelation estimates converge in about 20 to 40 iterations even for low signal-to-noise ratio.

An Intelligent Fault Diagnosis System for Machine Tools

An automatic intelligent system is developed to diagnose shaft fault types. Features related to shaft faults are extracted from vibration signals to effectively identify the corresponding fault condition. Feature extraction is accomplished using Fourier Transform, empirical mode decomposition (EMD) and multi-scale entropy (MSE). Through the EMD method, the model uses characteristics of intrinsic mode functions (such as zero-crossing rate and energy, to represent shaft condition features. MSE is used to calculate the entropy of multi-scale of the signal. At a larger MSE scale, the MSE result can be used to clearly identify some shaft defect types. The conventional approach to monitoring of a machine’s health online based on linear time-frequency analysis is subject to limitations, as the mechanical vibration signal is nonlinear and non-stationary in nature. Thus this research develops a diagnostic system based on the implementation of Fourier, EMD and MSE-based methods. In the buildup stage a knowledgeware is created from a database of existing defect types. Finally, the automatic intelligent monitoring system is implemented in a machine tool manufacturing company to verify its performance.

Measuring the effect of aging on vibrations of the carotid artery wall using empirical mode decomposition method

Quantification of arterial elasticity and its dependency to age is considered in this paper. We use radiofrequency (RF) signals from carotid artery ultasonography to evaluate this dependency. Blood pressure, blood flow, and tethering to surrounding tissue are the main causes of the motion of the carotid wall. Tracking carotid artery wall motion from a series of ultrasound B-mode images is challenging due to the presence of noise and variable contrast. Moreover, the process of converting RF signals into the B-mode images causes some information to be lost. Hence, our goal is to extract the carotid wall motions and vibrations from RF signals. After extraction and removing the wall motion by using the phased tracking method combined with continuous wavelet transform, the vibrations of carotid inner wall in different subjects in different ages are compared with each other. Empirical mode decomposition method is used for extracting the first intrinsic mode function for different subjects' vibration and then their zero-crossing rates are compared. The results show the vibrations of the carotid inner wall are clearly decreased by age.

Driver Voice Identification System Using Auto-Correlation Function and Average Magnitude Difference Function

This study presents a driver identification system using voice analysis for a vehicle security system. The structure of the proposed system has three parts. The first procedure is speech pre-processing, the second is feature extraction of sound signals, and the third is classification of driver voice. Initially, a database of sound signals for several drivers was established. The volume and zero-crossing rate (ZCR) of sound are used to detect the voice end-point in order to reduce data computation. Then the Auto-correlation Function (ACF) and Average Magnitude Difference Function (AMDF) methods are applied to retrieve the voice pitch features. Finally these features are used to identify the drivers by a General Regression Neural Network (GRNN). The experimental results show that the development of this voice identification system can use fewer feature vectors of pitch to obtain a good recognition rate.

Estimating the scattering distribution of the received signals in multipath fading channels

This paper presents a novel approach for estimating the distribution of the incoming waves at the mobile unit antenna, i.e. the scattering distribution, in a typical micro-cellular system. This estimate is vital in determining many system parameters of interest as well as designing unbiased estimators for the velocity of mobile units in micro-cellular systems. The proposed approach deploys the zero-crossing rates of the quadrature components and the instantaneous frequency of the received signal at the mobile unit to estimate the scattering distribution. We also propose a new model for simulating multipath fading channels with non-isotropic scattering. We use the channel simulator to evaluate the performance of the proposed estimator for the scattering distribution. Simulation results show that proposed estimator exhibits small bias and root mean square error.

Speech Emotion Recognition By Using Combinations Of C5.0, Neural Network (nn), And Support Vector Machines (svm) Classification Methods

Speech is the fastest and most natural method for human to communicate. This has led several researches to be done in the field of the interaction effects between human and machine. Hence, it is necessary to design machines which can intelligently recognize the emotion of a human voice. However, we are still far from having a natural interaction between the human and machine because machines cannot distinguish the emotion of the speaker. This has established a new field in the literature, namely the speech emotion recognition systems. The accuracy of these systems depends on various factors such as the number and type of the emotion manners as well as the feature selection and the classifier sort. In this paper, classification methods of the Neural Network (NN), Support Vector Machine (SVM), the combination of NN and SVM (NN-SVM), NN and SVM (NN-SVM), NN and C5.0 (NN-C5.0), C5.0 and SVM (SVM-C5.0), and finally the combination of NN, SVM, and C5.0 (NN-SVM-C5.0) have been verified, and their efficiencies in speech emotion recognition have been compared. The utilized features in this research include energy, power, Zero Crossing Rate (ZCR), pitch, and Mel-scale Frequency Cepstral Coefficients (MFCC). The presented results in this paper demonstrate that using the proposed NN-C5.0 classification method is more efficient in recognizing the emotion states-to the extent of 6%- to 30% depending on the number of emotions states-than SVM, NN, and other aforementioned combinations of classification methods.

Intelligent Adaptive Active Noise Control in Non-stationary Noise Environments

The famous filtered-x least mean square (FxLMS) algorithm for active noise control (ANC) systems may become unstable in non-stationary noise environment. To solve this problem, Sun`s algorithm and Akhtar`s algorithm are developed based on modifying the reference signal in update of FxLMS algorithm, but these two algorithms have dissatisfactory stability in dealing with sustaining impulsive noise. In proposed algorithm, probability estimation and zero-crossing rate (ZCR) control are used to improve the stability and performance, at the same time, an optimal parameter selection based on fuzzy system is utilized. Computer simulation results prove the proposed algorithm has faster convergence and better stability in non-stationary noise environment.

基于Matlab的语音信号时域分析系统

本文设计了一个语音信号分析处理系统，并在Matlab环境下进行了实现。该系统能够实现信号的采集与播放，同时对语音信号进行时域分析与处理，获得信号的短时过零率，短时能量、短时自相关函数等参数。该系统充分利用了Matlab强大的数值计算、信号处理、图形显示等能力，具有使用方便，通用性强等优点。 A speech signal analysis and processing system is realized in this paper under the Matlab environment. Signal acquisition and playback are also achieved in the system and a lot of time-domain parameters can be obtained while signal analysis is processing, such as short-term signal zero-crossing rate, short-time energy, short-time auto- correlation function, and so on. Powerful numerical calculation, signal processing, and graphics display capabilities of Matlab are fully used in this system. It is easy to use, and it has the versatility and other advantages.

Voice Activity Based on Noise Estimation in Noisy Environments

An improved project based on decision trees in noisy environments is proposed for robust endpoints detection. Firstly, the noise level of the environment is estimated by wavelet decomposition, and then whether the denoising process is done according to the noise level is determined. Next, the thresholds are obtained by decision trees for the signal. Finally, endpoints are detected by the double thresholds on different importance of the energy and zero-crossing rate (ZCR) in the corresponding situation. The simulation results indicate that the proposed method based on noise estimation can obtain the same accurate data by computing less than the one with decision trees.

Measurement of the Transient Velocity of Nanoparticles in Flowing Nanofluids by Using the Method of Dynamic Laser Speckles

Experimental equipments are designed for the zero-crossing method of dynamic laser speckles to investigate the moving nature of nanoparticles in flowing nanofluids. After the experiments the velocity of nanoparticles in flowing nanofluids is determined according to its relationship with the zero-crossing rate of speckle intensity fluctuation. The results show that the zero-crossing rate is very sensitive to the velocity of nanoparticles when the velocity of nanofluids ranges from 17.6 mm/s to 60.2 mm/s. A TEM10 mode–like laser beam is designed and taken as an incident light to detect the moving direction of nanoparticles in flowing nanofluids. The moving direction of nanoparticles is parallel to the horizontal axis of the TEM10 mode-like laser beam when the zero-crossing rate reaches maximum. Experimental results show that the moving direction of nanoparticles can also be detected by using the zero-crossing method.

Comparing Supervised Learning Techniques on the Task of Physical Activity Recognition

The objective of this study was to compare the performance of base-level and meta-level classifiers on the task of physical activity recognition. Five wireless kinematic sensors were attached to each subject (n = 25) while they completed a range of basic physical activities in a controlled laboratory setting. Subjects were then asked to carry out similar self-annotated physical activities in a random order and in an unsupervised environment. A combination of time-domain and frequency-domain features were extracted from the sensor data including the first four central moments, zero-crossing rate, average magnitude, sensor cross-correlation, sensor auto-correlation, spectral entropy and dominant frequency components. A reduced feature set was generated using a wrapper subset evaluation technique with a linear forward search and this feature set was employed for classifier comparison. The meta-level classifier AdaBoostM1 with C4.5 Graft as its base-level classifier achieved an overall accuracy of 95%. Equal sized datasets of subject independent data and subject dependent data were used to train this classifier and high recognition rates could be achieved without the need for user specific training. Furthermore, it was found that an accuracy of 88% could be achieved using data from the ankle and wrist sensors only.

Artificial neural networks for breathing and snoring episode detection in sleep sounds

Obstructive sleep apnea (OSA) is a serious disorder characterized by intermittent events of upper airway collapse during sleep. Snoring is the most common nocturnal symptom of OSA. Almost all OSA patients snore, but not all snorers have the disease. Recently, researchers have attempted to develop automated snore analysis technology for the purpose of OSA diagnosis. These technologies commonly require, as the first step, the automated identification of snore/breathing episodes (SBE) in sleep sound recordings. Snore intensity may occupy a wide dynamic range (>95 dB) spanning from the barely audible to loud sounds. Low-intensity SBE sounds are sometimes seen buried within the background noise floor, even in high-fidelity sound recordings made within a sleep laboratory. The complexity of SBE sounds makes it a challenging task to develop automated snore segmentation algorithms, especially in the presence of background noise. In this paper, we propose a fundamentally novel approach based on artificial neural network (ANN) technology to detect SBEs. Working on clinical data, we show that the proposed method can detect SBE at a sensitivity and specificity exceeding 0.892 and 0.874 respectively, even when the signal is completely buried in background noise (SNR <0 dB). We compare the performance of the proposed technology with those of the existing methods (short-term energy, zero-crossing rates) and illustrate that the proposed method vastly outperforms conventional techniques.