Microphone Sensor Research Articles

An acoustic system for EOL engines diagnoses in hot test cells

End of Line tests have a crucial role in industrial processes to validate the products’ quality. In automotive industry, engines at the end of assembling process are fully tested within hot test cells at different working conditions by mean specific test cycles. Vibration monitoring is widely used in order to identify potential faults in the assembly process. This contact technique leads to several disadvantages, mainly due to the accessibility and the characteristics of surfaces for accelerometers mounting (e.g. rotating parts, geometries complexes, high temperatures) and to the time-consuming setup procedures. These limitations can be overcome by using microphone sensors. This paper describes the development phases of an acoustic based fault diagnosis system, its integration on hot test cells and the results obtained over large number of engines tested. By mean some specific technical solutions the acoustics approach in the EOL fault diagnosis of combustion engines has showed itself to be compatible with the characteristics of the production industry environment, providing reliable and repeatable results. The latter has been demonstrated also by comparison with the traditional accelerometer technique.

Nondestructive Detection of Gaps between Railway Track Slabs and Soil Foundation Using Leaked Air Waves

Gaps generated underneath railway track slabs may cause unstable conditions. Such gaps can form because of different reasons, including settlement of soft soil, unsuitable construction of a concrete slab on loosely compacted soil, and drastic stiffness change in a transition zone between the bridge deck and embankment. The gaps underneath railway track slabs are not easily detectable by common nondestructive test methods. A nondestructive test (NDT) based on a wavelet time–frequency concept is proposed for the practical purpose of detecting gaps under track slabs. The method uses a microphone sensor to catch leaked Rayleigh acoustic waves in the air and an accelerometer to measure surface Rayleigh waves on the slab. In order to investigate the possibility of developing the test system, a finite element analysis (FEA) was performed to simulate Rayleigh wave generation on the concrete surface and in the air. A test system module composed of a microphone and an accelerometer, data acquisition system (DAQ), and an analyzer program was also assembled for a small backyard pilot test. It was verified that the new NDT test system could be successfully adapted for detecting gaps underneath railway track slabs and track bed soil.

Acoustic MEMS Sensor Array for Quench Detection of CICC Superconducting Cables

A novel quench detection method using micro-electro-mechanical system (MEMS) sensor technology has been investigated in use for high temperature superconducting (HTS) conductors such REBCO tape cables. The sensor array along a superconducting cable, such as a cable-in-conduit-conductor (CICC), is installed in a cooling channel. It will allow sensitive and quick detection for a local quench of a superconducting cable. This work has confirmed that a quench of a single REBCO tape can be detected in liquid nitrogen by a MEMS piezoelectric microphone sensor. The quench detection design utilizing a MEMS sensor array method is discussed for the case of a toroidal field (TF) magnets of a fusion Tokamak device.

Improved Empirical Wavelet Transform for Compound Weak Bearing Fault Diagnosis with Acoustic Signals

Most of the current research on the diagnosis of rolling bearing faults is based on vibration signals. However, the location and number of sensors are often limited in some special cases. Thus, a small number of non-contact microphone sensors are a suboptimal choice, but it will result in some problems, e.g., underdetermined compound fault detection from a low signal-to-noise ratio (SNR) acoustic signal. Empirical wavelet transform (EWT) is a signal processing algorithm that has a dimension-increasing characteristic, and is beneficial for solving the underdetermined problem with few microphone sensors. However, there remain some critical problems to be solved for EWT, especially the determination of signal mode numbers, high-frequency modulation and boundary detection. To solve these problems, this paper proposes an improved empirical wavelet transform strategy for compound weak bearing fault diagnosis with acoustic signals. First, a novel envelope demodulation-based EWT (DEWT) is developed to overcome the high frequency modulation, based on which a source number estimation method with singular value decomposition (SVD) is then presented for the extraction of the correct boundary from a low SNR acoustic signal. Finally, the new fault diagnosis scheme that utilizes DEWT and SVD is compared with traditional methods, and the advantages of the proposed method in weak bearing compound fault diagnosis with a single-channel, low SNR, variable speed acoustic signal, are verified.

A CNN-Assisted Enhanced Audio Signal Processing for Speech Emotion Recognition.

Speech is the most significant mode of communication among human beings and a potential method for human-computer interaction (HCI) by using a microphone sensor. Quantifiable emotion recognition using these sensors from speech signals is an emerging area of research in HCI, which applies to multiple applications such as human-reboot interaction, virtual reality, behavior assessment, healthcare, and emergency call centers to determine the speaker’s emotional state from an individual’s speech. In this paper, we present major contributions for; (i) increasing the accuracy of speech emotion recognition (SER) compared to state of the art and (ii) reducing the computational complexity of the presented SER model. We propose an artificial intelligence-assisted deep stride convolutional neural network (DSCNN) architecture using the plain nets strategy to learn salient and discriminative features from spectrogram of speech signals that are enhanced in prior steps to perform better. Local hidden patterns are learned in convolutional layers with special strides to down-sample the feature maps rather than pooling layer and global discriminative features are learned in fully connected layers. A SoftMax classifier is used for the classification of emotions in speech. The proposed technique is evaluated on Interactive Emotional Dyadic Motion Capture (IEMOCAP) and Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) datasets to improve accuracy by 7.85% and 4.5%, respectively, with the model size reduced by 34.5 MB. It proves the effectiveness and significance of the proposed SER technique and reveals its applicability in real-world applications.

Closed loop force control of in-situ machining robots using audible sound features

Abstract Detecting, measuring and controlling the forces between cutting tools and machined components is essential in circumstances where direct position control (e.g. depth of cut, feed speed, etc.) is inaccurate and/or impossible. This paper explores the use of airborne sound signals that result from the machining process to control the cutting force in closed loop for generating accurate machined features when performing in-situ robotic repair of complex installations. The sound signals during indentation at various cutting forces are analysed and used to calibrate a remotely mounted microphone sensor and signal processing control system. The power spectral density of the audible sound is used to estimate tool cutting force and the sound intensity used in turn to estimate the resulting process energy. The described controller uses intensity of sound to mitigate the effects of resonance with workpiece natural frequencies while controlling the spindle velocity of the tool based on the dominant audible frequency. The performance of the controller is validated using a representative test rig and demonstrated using a robotic arm to machine thin Ni-Cr-Co alloy cantilever beams with a miniature air-driven grinding tool. Results from the test rig show that such a sound-based control approach can achieve consistent cutting forces with an accuracy of 0.08 N. The robot arm is shown to be capable of grinding features of consistent depth (to within 0.05 mm) on beams with surface defects of undefined shape using only the sound of the process for closed loop force control.

Implementation of Artificial Intelligence for Classification of Frogs in Bioacoustics

This research presents the implementation of artificial intelligence (AI) for classification of frogs in symmetry of the bioacoustics spectral by using the feedforward neural network approach (FNNA) and support vector machine (SVM). Recently, the symmetry concept has been applied in physics, and in mathematics to help make mathematical models tractable to achieve the best learning performance. Owing to the symmetry of the bioacoustics spectral, feature extraction can be achieved by integrating the techniques of Mel-scale frequency cepstral coefficient (MFCC) and mentioned machine learning algorithms, such as SVM, neural network, and so on. At the beginning, the raw data information for our experiment is taken from a website which collects many kinds of frog sounds. This in fact saves us collecting the raw data by using a digital signal processing technique. The generally proposed system detects bioacoustic features by using the microphone sensor to record the sounds of different frogs. The data acquisition system uses an embedded controller and a dynamic signal module for making high-accuracy measurements. With regard to bioacoustic features, they are filtered through the MFCC algorithm. As the filtering process is finished, all values from ceptrum signals are collected to form the datasets. For classification and identification of frogs, we adopt the multi-layer FNNA algorithm in machine learning and the results are compared with those obtained by the SVM method at the same time. Additionally, two optimizer functions in neural network include: scaled conjugate gradient (SCG) and gradient descent adaptive learning rate (GDA). Both optimization methods are used to evaluate the classification results from the feature datasets in model training. Also, calculation results from the general central processing unit (CPU) and Nvidia graphics processing unit (GPU) processors are evaluated and discussed. The effectiveness of the experimental system on the filtered feature datasets is classified by using the FNNA and the SVM scheme. The expected experimental results of the identification with respect to different symmetry bioacoustic features of fifteen frogs are obtained and finally distinguished.

Prototype Sistem Deteksi Partial Discharge Pada Isolasi Kabel Menggunakan Sensor Microphone

The phenomenon that often arises in a substation is the problem of partial discharge in outgoing cable insulation. Partial discharge is a jump of positive and negative ions that are not supposed to meet so that it can cause a spark jump. If a partial discharge is left too long it can cause insulation failure, the sound of snakes like hissing and the most can cause a flashover on the outgoing cable. Then a partial discharge detection prototype was made in the cable insulation in order to anticipate the isolation interference in the outgoing cable. Can simplify the work of substation operators to check the reliability of insulation on the outgoing side of each cubicle. So it was compiled as a method for measuring sound waves caused by partial discharge in the process of measuring using a microphone sensor, the Arduino Mega 2560 module as a microcontroller, the LCD TFT as a monitoring and the MicroSD card module as its storage. The microphone sensor is a sensor that has a high sensitivity to sound, has 2analog and digital readings, and is easily designed with a microcontroller. Basically the unit of measure measured at partial discharge is Decibels. The results of the prototype can be applied to the cubicle and the way it works is to match the prototype to the outgoing cubicle cable then measure from the cable boots connector to the bottom of the outgoing cable with a distance of 1 meter. Then the measurement results will be monitored on the TFT LCD screen in the form of measurement results, graphs and categories on partial discharge. In this design the measurement data made by the microphone can be stored with microSD so that it can make an evaluation of partial discharge handling in outgoing cable insulation.

Design and Implementation of Heart Sound Detection Device Based on MEMS MIC

The paper describes how to develop a digital heart sound signal detection device based on high gain MEMS MIC that can accurately collect and store human heart sounds. According to the method of collecting heart sound signal by traditional stethoscope, the system improves the traditional stethoscope, and a composite probe equipped with a MEMS microphone sensor is designed. The MEMS microphone sensor converts the sound pressure signal into a voltage signal, and then amplifies, converts with Sigma Delta, extracts and filters the collected signal. After the heart sound signal is uploaded to the PC, the Empirical Mode Decomposition (EMD) is carried out to reconstruct the signal, and then the Independent Component Analysis (ICA) method is used for blind source separation and finally the heart rate is calculated by autocorrelation analysis. At the end of the paper, a preliminary comparative analysis of the performance of the system was carried out, and the accuracy of the heart sound signal was verified.

A VCO-Based CMOS Readout Circuit for Capacitive MEMS Microphones.

Microelectromechanical systems (MEMS) microphone sensors have significantly improved in the past years, while the readout electronic is mainly implemented using switched-capacitor technology. The development of new battery powered “always-on” applications increasingly requires a low power consumption. In this paper, we show a new readout circuit approach which is based on a mostly digital Sigma Delta () analog-to-digital converter (ADC). The operating principle of the readout circuit consists of coupling the MEMS sensor to an impedance converter that modulates the frequency of a stacked-ring oscillator—a new voltage-controlled oscillator (VCO) circuit featuring a good trade-off between phase noise and power consumption. The frequency coded signal is then sampled and converted into a noise-shaped digital sequence by a time-to-digital converter (TDC). A time-efficient design methodology has been used to optimize the sensitivity of the oscillator combined with the phase noise induced by and thermal noise. The circuit has been prototyped in a 130 nm CMOS process and directly bonded to a standard MEMS microphone. The proposed VCO-based analog-to-digital converter (VCO-ADC) has been characterized electrically and acoustically. The peak signal-to-noise and distortion ratio (SNDR) obtained from measurements is 77.9 dB-A and the dynamic range (DR) is 100 dB-A. The current consumption is 750 A at 1.8 V and the effective area is 0.12 mm. This new readout circuit may represent an enabling advance for low-cost digital MEMS microphones.

A method for estimating the amplitude response of smartphone built-in microphone sensors below 4 kHz.

A method for estimating the amplitude response of built-in smartphone microphone sensors is presented. The method is intended to be accessible to the general public, and comparison calibration measurements are performed in a non-isolated environment using conventional consumer products. A double reference sensor scheme is set up, consisting of a MB3 digital microbarometer (reference over the 0.5-2 Hz octave bands) and the iPrecisionMic smartphone microphone (reference above the 2 Hz octave band). The amplitude response of the iPrecisionMic sensor is first evaluated over the 1-2000 Hz octave bands. The amplitude response of three Samsung Galaxy S8 built-in smartphone microphone sensors is then measured over the 0.5-2000 Hz octave bands. The Redvox Infrasound Recorder application (app) for Android is used to measure acoustic signals with the built-in smartphone microphone sensors. Amplitude response models in terms of digital gain are estimated for the test sensors based on the results. Last, self-noise levels for the iPrecisionMic and Samsung Galaxy S8 microphones are estimated and compared to infrasound ambient noise models. Results show an experimental capability for estimating the amplitude response of built-in smartphone microphone sensors in a non-isolated environment with conventional consumer products.

A multi-sensor information fusion for fault diagnosis of a gearbox utilizing discrete wavelet features

A gear box is widely employed in automobiles and industrial machines for transmitting power and torque. It operates under various working conditions for prolonged hours increasing the chance of gearbox failure. Major faults in gear systems are caused due to wear, scoring, pitting, tooth fracture, etc. Gear box failure leads to increases in machine downtime and maintenance costs. The nature and location of such failures can be identified with precision using condition monitoring techniques. In this study, machine condition data are acquired from the gear box using a vibration accelerometer, microphone and acoustic emission sensors under different operating conditions, such as three loading conditions (0 N, 5 N, 10 N) and three rpm variations (500, 750, 1000). The wavelet features are extracted from the acquired vibration, sound and acoustic emission signal, and prominent features are identified. To automate the process of fault diagnosis, machine learning algorithms (artificial neural network, support vector machine, and proximal support vector machine) are utilized. Dual and multi-sensor fusion is implemented with the help of prominent features, to intensify the classification accuracy. The performance of the individual signals, and dual and multi-sensor fused models in gearbox fault diagnosis are compared and discussed in detail.

Multilevel B-Splines-Based Learning Approach for Sound Source Localization

In this paper, a new learning approach for sound source localization is presented using ad hoc either synchronous or asynchronous distributed microphone networks based on the time differences of arrival (TDOA) estimation. It is first to propose a new concept in which the coordinates of a sound source location are defined as the functions of TDOAs, computing for each pair of microphone signals in the network. Then, given a set of pre-recorded sound measurements and their corresponding source locations, the multilevel B-splines-based learning model is proposed to be trained by the input of the known TDOAs and the output of the known coordinates of the sound source locations. For a new acoustic source, if its sound signals are recorded, the correspondingly computed TDOAs can be fed into the learned model to predict the location of the new source. Superiorities of the proposed method are to incorporate the acoustic characteristics of a targeted environment and even remaining uncertainty of TDOA estimations into the learning model before conducting its prediction and to be applicable for both synchronous or asynchronous distributed microphone sensor networks. The effectiveness of the proposed algorithm in terms of localization accuracy and computational cost in comparisons with the state-of-the-art methods was extensively validated on both synthetic simulation experiments as well as in three real-life environments.

Nested sensor array extension factors required to match the peak sidelobe height of a uniform linear array

Nested sensor arrays (NSAs) are a type of non-uniform linear array (NULA) that reduce the total number of sensors used for a given aperture and retain the same resolution of a uniform linear array (ULA) [Pal and Vaidyanathan, 2010]. The peak sidelobe (PSL) height fails to fall below the acceptable standard of -13 dB associated with ULAs for basic NSA configurations. The basic NSA can be extended by a factor to improve the PSL height. Coprime sensor arrays, a related NULA design, have established sensor extension factors that have been found to reduce the PSL height to this -13 dB standard [Adhikari et al., 2014]. For NSAs, such an extension factor has not been established. This research finds the parameters required to extend an NSA to adequately reduce the PSLs. NSAs can significantly reduce the number of sensors required to yield a satisfactory beampattern, as low as twenty-five percent of the total aperture for some cases. Considering various array apertures, certain configurations successfully use a distinctly small aperture but require a higher percentage of those to be active. To verify our results, we construct a sixty-four sensor microphone array with which we test the optimal configurations. [Work supported by Louisiana Tech University.]

Smartphones Identification Through the Built-In Microphones With Convolutional Neural Network

The use of mobile phones or smartphones has become so widespread that most people rely on them for many services and applications like sending e-mails, checking the bank account, accessing cloud platforms, health monitoring, buying on-line and many other applications where sharing sensitive data is required. As a consequence, security functions are important in the use of smartphones, especially because most of the applications require the identification and authentication of the device in mobility. This is usually achieved through cryptographic systems but recent research studies have also investigated alternative or complementary authentication mechanisms which can be used to strengthen cryptographic methods with multi-factor authentication. In this paper, we investigate the identification and the authentication of smartphones using the intrinsic physical properties of the mobile phones built-in microphones. The possibility to identify a microphone on the basis of features extracted from audio recordings is well known in literature but it is mostly used in forensics studies and usually relies on human voice recordings. On the contrary this paper proposes a smartphone identification and authentication approach by stimulating the built-in microphone with non-voice sounds at different frequencies. An extensive data set of 32 phones was used to evaluate experimentally the proposed approach. On the basis of the proven performance of deep learning techniques, a new Convolutional Neural Network architecture is proposed both for the identification and the authentication purposes. Its performance, in comparison to other machine learning algorithms, is demonstrated in presence of different types of noises (e.g., Gaussian White noise, Babble noise and Street noise). Satisfactory results have been obtained showing that the exploitation of a fingerprint from the microphone sensor is a good choice to assess smartphone distinctiveness.

Portable multi-parameter electrical impedance tomography for sleep apnea and hypoventilation monitoring: feasibility study

Objective: Quantitative ventilation monitoring and respiratory event detection are needed for the diagnosis of sleep apnea and hypoventilation. We developed a portable device with a chest belt, nasal cannula and finger sensor to continuously acquire multi-channel signals including tidal volume, nasal pressure, respiratory effort, body position, snoring sound, ECG and SpO2. The unique feature of the device is the continuous tidal volume signal obtained from real-time lung ventilation images produced by the electrical impedance tomography (EIT) technique. Approach: The chest belt includes 16 electrodes for real-time time-difference EIT imaging and ECG data acquisitions. It also includes a microphone, accelerometer, gyroscope, magnetometer and pressure sensor to acquire, respectively, snoring sound, respiratory effort, body position and nasal pressure signals. A separate finger sensor is used to measure SpO2. The minute ventilation signal is derived from the tidal volume signal and respiration rate. Main results: The experimental results from a conductivity phantom, four swine subjects and one human volunteer show that the developed multi-parameter EIT device could supplement existing polysomnography (PSG) and home sleep test (HST) devices to improve the accuracy of sleep apnea diagnosis. The portable device could be also used as a new tool for continuous hypoventilation monitoring of non-intubated patients with respiratory depression. Significance: Following the feasibility study in this paper, future validation studies in comparison with in-lab PSG, HST and end-tidal CO2 devices are suggested to find its clinical efficacy as a sleep apnea diagnosis and hypoventilation monitoring tool.

Fetal heart rate monitoring device using condenser microphone sensor: Validation and comparison to standard devices

Fetal heart rate (FHR) monitoring device is highly demanded to assess the fetus health condition in home environments. Conventional standard devices such as ultrasonography and cardiotocography are expensive, bulky and uncomfortable and consequently not suitable for long-term monitoring. Herein, we report a device that can be used to measure fetal heart rate in clinical and home environments. The proposed device measures and displays the FHR on a screen liquid crystal display (LCD). The device consists of hardware that comprises condenser microphone sensor, signal conditioning, microcontroller and LCD, and software that involves the algorithm used for processing the conditioned fetal heart signal prior to FHR display. The device's performance is validated based on analysis of variance (ANOVA) test. FHR data was recorded from 22pregnant women during the 17th to 37th week of gestation using the developed device and two standard devices; AngelSounds and Electronic Stethoscope. The results show that F-value (1.5) is less than F𝑐𝑟𝑖𝑡, (3.1) and p-value (p> 0.05). Accordingly, there is no significant difference between the mean readings of the developed and existing devices. Hence, the developed device can be used for monitoring FHR in clinical and home environments.

Data mining of a clean signal from highly noisy data based on compressed data fusion: A fast-responding pressure-sensitive paint application

A data mining approach based on compressed data fusion is developed to extract a clean signal from highly noisy data and it has been successfully applied to flow measurement using fast-responding pressure-sensitive paint (fast PSP). In this approach, spatially resolved but noisy full-field data are fused with clean but scattered data to reconstruct full-field clean data. The fusion process is accomplished based on a compressed sensing algorithm, which has shown significantly improved performance compared with low-dimensional analysis. This is because, in low-dimensional analysis such as proper orthogonal decomposition (POD), the selection criteria of proper POD modes for reconstruction are usually based on subjective observation and the mode coefficients can be severely distorted by noise, which restricts the applications of this method to complicated flow phenomena and leads to a low-quality reconstruction. The solutions to these two problems can be expressed via mathematical optimization by determining the optimal coefficients to reconstruct clean data using the most relevant POD modes. Here, compressed sensing is used as a suitable solution to explore the sparse representation of scattered clean data based on the POD modes obtained from noisy full-field data. A high-quality reconstruction can be obtained using the optimized coefficients. The new method is first demonstrated by using fabricated patterns, demonstrating a reduction of 75% in the reconstruction error compared with POD analysis. It is thereafter successfully applied to recover the unsteady pressure field induced by a cylinder wake flow at low speed. Fast PSP measurement and microphones are used to obtain full-field but noisy pressure field data and scattered but clean data, respectively. In the cases of single and step cylinders, the reconstruction errors are approximately 5% and 25%, respectively, and the accuracy of reconstruction depends on the low dimensionality of the flow phenomena and the total number of microphone sensors. The current technique provides a reliable method to recover clean signals from strong noise, with significant potential for applications to flow measurement, control, and monitoring.

Validated data mining technique: a case study on pressure-field digital MEMS microphone calibration

Recently, in the field of environmental noise control and monitoring, a diffusion of a new generation of microphone sensors has been observed. This was met by the rapid development of MEMS (MicroElectroMechanical Systems) microphones, whose technology could open up the possibility of noise measurements based on extensive acoustic sensor networks, that use a large number of microphones spread across wide areas. Nevertheless, current standards do not provide proper calibration and test procedures for these microphones. In this work, calibration standard procedures was adapted for the complete characterization of digital MEMS microphones by comparison with laboratory standard microphones in pressure field condition. Large data-set obtained from comparison, is subjected to a statistical analysis in order to highlight any possible factor affecting calibration accuracy and robustness.

Design and Implementation of Acoustic Sensing System for Online Early Fault Detection in Industrial Fans

Industrial fans play a critical role in manufacturing facilities, and a sudden shutdown of critical fans can cause significant disruptions. Ensuring early, effective, and accurate detection of fan malfunctions first requires confirming the characteristics of anomalies resulting from initial damage to rotating machinery. In addition, sensing and detection must rely on the use of sensors and sensing characteristics appropriate to various operational abnormalities. This research proposes an online industrial fan monitoring and fault detection technique based on acoustic signals as a physical sensing index. The proposed system detects and assesses anomalies resulting from preliminary damage to rotating machinery, along with improved sensing resolution bandwidth features for microphone sensors as compared to accelerometer sensors. The resulting Intelligent Prediction Integration System with Internet (IPII) is built to analyze rotation performance and predict malfunctions in industrial fans. The system uses an NI cRIO-9065 embedded controller and a real-time signal sensing module. The kernel algorithm is based on an acoustic signal enhancement filter (ASEF) as well as an adaptive Kalman filter (AKF). The proposed scheme uses acoustic signals with adaptive order-tracking technology to perform algorithm analysis and anomaly detection. Experimental results showed that the acoustic signal and adaptive order analysis method could effectively perform real-time early fault detection and prediction in industrial fans.