Signal Classification Algorithm Research Articles

Vehicle Acoustic and Seismic Synchronization Signal Classification Using Long-Term Features

Field vehicle type recognition plays an essential role in border protection tasks. Acoustic and seismic sensors can effectively collect the signal of field vehicle targets in real time. Most vehicle temporal signal classification algorithms are based on extracting and identifying handcrafted features. These algorithms focus on the signal’s frequency domain characteristics and despise the signal’s time domain characteristics. In order to extract appropriate features, this paper proposes a Long-Term Correlation Feature Network (LTCFN) to perform field vehicle acoustic and seismic signal classification. The model includes AlexNet-type feature extractor and an overall classifier implemented by a Long-Short Term Memory(LSTM) network. We presents an intra-frame network and fusion method for extracting feature vector from signals. Meanwhile, an inter-frame classifier is proposed firstly for analysing the time-correlation of the feature map and overall classification. The experiments illustrate that the LTCFN has excellent recognition performance and anti-noise ability. The classification accuracy of the LTCFN can be increased to 96%. This paper also provides a new idea for ground target classification through inter-frame feature measurement.

Cross Dipole Antennas Solution for Angle of Arrival Estimation

The Multiple Signal Classification (MUSIC) algorithm is the most popular algorithm to estimate the Angle of Arrival (AOA) of the received signals. The analysis of this algorithm (MUSIC) with typical array antenna element ( ) shows that there are two false direction indication in the planaligned with the axis of the array. In this paper a suggested modification on array system is proposed by using two perpendiculars crossed dipole array antenna in spite of one array antenna. The suggested modification does not affect the AOA estimation algorithm. The simulation and results shows that the proposed solution overcomes the MUSIC problem without any effect on the performance of the system.

A Novel Wide-Band Directional MUSIC Algorithm Using the Strength Proportion.

The directional multiple signal classification (Dir-MUSIC) algorithm based on the antenna gain array manifold has been proposed to find the direction of the partial discharge (PD) source in substations. However, PD signals are wideband signals and the antenna gain pattern functions are always different at different frequencies; therefore, the accuracy can be improved using a wideband Dir-MUSIC algorithm. In this paper, wideband Dir-MUSIC algorithms are discussed and a novel wideband Dir-MUSIC algorithm using the strength proportion (DirSP) is proposed. This algorithm estimates a focusing PD signal at a certain frequency using the strength proportion among different directions, and then the Dir-MUSIC algorithm can process the focusing PD signal at this frequency. In simulations, when the antenna gain functions among different frequency bins are quite different, the Dir-MUSIC algorithm loses accuracy; meanwhile, DirDP performs very well. In the experiments, we deal with six sets of samples, and the mean error and standard deviation are both smaller than 4° better than other methods.

Multi-Leakage Source Localization of Safety Valve Based on Uniform Circular AE Array and Improved MUSIC Algorithm.

The safety valve is the core component of the pressure-relief protection device for pressure-bearing special equipment. When the safety valve leaks, the medium of the pressure vessel will be lost and wasted, which may cause safety accidents. With the aim to solve the problem of accurately locating the multiple leakage sources of safety valves, a localization method combining a uniform circular array acoustic emission detection and an improved multiple signal classification (MUSIC) algorithm is proposed. First, an improved wavelet threshold function denoising method is introduced to extract acoustic emission signals with high SNR, thereby reducing the rank of the covariance matrix, weakening the noise dispersion caused by eigenvalue reconstruction, avoiding signal and noise cross-confusion, and improving positioning accuracy. By introducing a windowed fast Fourier transform (FFT) frequency division processing link to obtain narrowband signal, the premise of using MUSIC positioning algorithm is established. In addition, a forward/backward spatial smoothing algorithm is introduced in the decoherence link to reduce co-channel interference, reduce the rank loss of the signal covariance matrix, and improve the positioning accuracy of the algorithm. The results show that when the working pressure is 0.70 MPa, 0.75 MPa, and 0.80 MPa, the deviation between the azimuth angle and elevation angle positioning results of each leakage source obtained by the improved MUSIC algorithm and the actual angle does not exceed 2°, and the relative error does not exceed 3.5%. Therefore, the improved MUSIC algorithm can accurately locate multiple leakage sources of the safety valve, and as the working pressure of the safety valve increases, the positioning accuracy of the improved MUSIC algorithm also increases accordingly.

High-Performance of Eigenvalue Decomposition on FPGA for the DOA Estimation

For the direction of arrival (DOA) in array signal processing, eigenvalue decomposition (EVD) is one key issue in hardware implementation of the multiple signal classification (MUSIC) algorithm. Therefore, we introduce the look-ahead sim- plified one-sided Jacobi's method to efficiently decompose those symmetric matrices in this article and prove that the new method has the best orthogonality of eigenvector and locates eigenvectors closest to the true solution in theory. Both the numerical perform- ance and real-time are important in engineering, so we present the novel flexible hardware architecture in single floating point arithmetic for EVD on field-programmable gate arrays (FPGAs). Finally, the simulated and raw data are used to investigate the performance of some different approaches in the context of both the EVD and MUSIC algorithm. The experimental results show that our proposed method has the best performance.

A Double Fourier-Transform Imaging Algorithm for a 24 GHz FMCW Short-Range Radar.

A frequency-modulated continuous-wave radar for short-range target imaging, assembling a transceiver, a PLL, an SP4T switch, and a serial patch antenna array, was realized. A new algorithm based on a double Fourier transform (2D-FT) was developed and compared with the delay and sum (DAS) and multiple signal classification (MUSIC) algorithms proposed in the literature for target detection. The three reconstruction algorithms were applied to simulated canonical cases evidencing radar resolutions close to the theoretical ones. The proposed 2D-FT algorithm exhibits an angle of view greater than 25° and is five times faster than DAS and 20 times faster than the MUSIC one. The realized radar shows a range resolution of 55 cm and an angular resolution of 14° and is able to correctly identify the positions of single and multiple targets in realistic scenarios, with errors lower than 20 cm.

An efficient damage imaging method for composite structure based on self-correcting phase error compensation MUSIC algorithm

The increasing use of composite materials on aircraft structures, most of which are plate-like, has attracted much attention for damage monitoring based on guided wave as a kind of structural health monitoring (SHM) method. The multiple signal classification (MUSIC) algorithm is a directional scanning and searching method to unbiasedly estimate the signal features in terms of the orthogonal attributes between signal subspace and noise subspace. The MUSIC algorithm is gradually shifted into the field of SHM used for damage localization. However, the anisotropy of the composites and the sensor position errors cause the phase errors of the array response signal, resulting in a decrease in the accuracy of the MUSIC method. In addition, the complexity of the array monitoring network causes a significant increase in the amount of operations for signal information processing, making online real-time monitoring challenging. In order to reduce the impact of multiple forms of array errors on the MUSIC algorithm and to improve the efficiency of it, this paper proposes an efficient damage imaging method for composite structure based on self-correcting phase error compensation MUSIC algorithm, which compensates for the array phase errors caused by structural anisotropy and sensor position errors, and achieves high efficiency. The method has been verified on a stiffened composite structure, and the results show its superiority and effectiveness.

High-Performance Channel Estimation for mmWave Wideband Systems With Hybrid Structures

In this paper, a channel estimation problem for millimeter-wave (mmWave) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems with hybrid structures is studied. Firstly, a beamspace multiple signal classification (MUSIC) algorithm for mmWave wideband channels is proposed to simultaneously estimate the angles of arrival (AOAs), angles of departure (AODs) and transmission delays. Since the traditional spectral peak search method has high complexity, a multi-spectral peak search method is skillfully designed to search for multiple spectral peaks on the MUSIC spatial spectrum more quickly and accurately. Then, the proposed channel estimator is extended to more actual systems equipped with uniform planar arrays (UPAs). Finally, the Cramér–Rao bound (CRB) results of these channel parameters are derived for evaluating the performance of the proposed channel parameter estimator. Simulation results demonstrate that the proposed channel estimator has greatly high channel estimation accuracy.

SEISMIC WAVE RECORDER DESIGN FOR DEVELOPMENT FOOTSTEP HUMAN DETECTION ALGORITHMS

The article shoving the results of development seismic signal recorder for purpose of registration human footstep seismic signals. Design of recorder was adopted for collection seismic signals and store it on PC, for next seismic signal processing and creation signal detection and classification algorithms.
 The article provides an overview of the types of seismic waves, their classification is given. The conditions for the propagation of seismic waves are determined, and the features of Rayleigh waves are clarified. The design structure of the seismic sensor has been developed. The specifics of the functioning of the seismic sensor are described, the place and conditions of power supply and placement of the external interface are recognized. The specifics of using the sensor of an electromechanical transducer - geophone and the features of measurements using a stationary earth are described. It is shown that the development of a seismic signal sensor circuit implies the placement of input amplifiers in its structure. The main amplitude-frequency and phase-frequency characteristics of the geophone are presented. Geophone equivalent circuit, amplifier circuit, low-pass filter circuit are presented. Illustrations of types of seismograms are given, recommendations on the features of digital signal processing are given. The order of seismic signal processing is connected with its transmission from the sensor through the amplifying stage, passing through a low-pass filter in order to minimize noise and interference. The next step is to send a signal to the ADC. From the presented graphs of sensitivity and phase shift of the geophone, it follows that special attention should be paid to the curve corresponding to the sensor with an open output. This configuration of the sensor is optimal in the context of the formation of the maximum equal frequency response.
 As papers result was proposed structural and schematic design for all main recorders blocks and software features that required for properly using recorder. Tests in real conditions on test sites have shown the correctness of the choice of schemes of technical solutions for the task of registering seismic waves.

Forward-propagation-free focusing MUSIC algorithm for Lamb waves

The multiple signal classification (MUSIC) algorithm is originally used for direction of arrival estimation and frequency identification. MUSIC-like algorithms have gained their popularity in structural health monitoring (SHM) owing to their super-resolution characteristic and the virtue of baseline independence. MUSIC essentially is a searching algorithm to find the best phase difference matching between the eigenvector from signal subspace and the presupposed steering vector. However, in the guided-wave-based SHM (in plate-like structure, the guided wave is Lamb wave), the dispersion effect will definitely lead to phase distortion. Therefore, Xu et al. presented a focusing MUSIC algorithm where the time reversal technique is adopted to eliminate the dispersion in the way of back propagation. Seeing that the phase difference information, which is necessary for MUSIC, disappears after back propagation, the damage scattered wave packet, obtained by window truncation, is forward propagated again at the constant group velocity to recover the time of flight information. Nevertheless, the forward propagation process is superfluous from the view of phase matching. In the presented forward-propagation-free focusing MUSIC (FPF-F MUSIC) algorithm (the simplified version of Xu’s work), by revising the steering vector in the form of phase coincidence, the matching relation for MUSIC is modified from phase difference matching to phase alignment matching, so that the forward propagation and the steering vector traversal setting for every inspection point are all omitted. The damage localization performance of FPF-F MUSIC is verified on the aluminum plate with double through-hole and surface defects under uniform linear and circular arrays.

Waveform design for higher-resolution localization with MIMO radar

To improve the accuracy and resolution of angle estimation with multiple signal classification (MUSIC) algorithm, we study the design of low peak-to-average-power-ratio waveforms for multiple-input-multiple-output (MIMO) radar. Assuming that the prior knowledge about the target directions and amplitudes is available (e.g., by transmitting initial probing waveforms), we formulate a waveform design problem based on minimizing the asymptotic mean square error of MUSIC. An alternating direction method of multipliers is developed to tackle the encountered nonconvex optimization problem. Compared with the competing waveforms, the waveforms designed by the proposed algorithm exhibit higher resolution and lower estimation error.

Joint DOA, range and polarization estimation based on geometric algebra for near-field non-circular source with a symmetric MIMO array

Multiple-input multiple-output (MIMO) radar is a new type of radar with excellent performance in target detection and parameter estimations. The MIMO radar equipped with electromagnetic vector sensors (EMVSs) can record multi-dimensional information of the electromagnetic (EM) signal. Traditional methods based on long vector model cannot preserve orthogonality between the outputs of each component of the EMVS and have high computational complexity. In this paper, a novel geometric algebra (GA)-based localization method is proposed for direction of arrival (DOA), range, and polarization estimation of near-field (NF) non-circular (NC) sources for a symmetric MIMO array. First, the proposed method converts the output of an EMVS into a multi-vector through a GA matrix transformation. Then, the DOA and range parameters can be estimated by searching the one-dimensional (1-D) spectrum based on the GA-based NC rank reduction (GANC-RARE) algorithm. Finally, the polarization can be estimated by employing estimated DOAs and ranges based on the modified GA-based NC multiple signal classification (MGANC-MUSIC) algorithm. The computational complexity analysis demonstrates that our solution improves memory spaces of the covariance matrix and reduces computation efforts of the eigenvalue decomposition (EVD). The simulation results show the superiority and reliability of our proposed method in model error and coherent noise environment.

DOA Estimation of Indoor Sound Sources Based on Spherical Harmonic Domain Beam-Space MUSIC

The Multiple Signal Classification (MUSIC) algorithm has become one of the most popular algorithms for estimating the direction-of-arrival (DOA) of multiple sources due to its simplicity and ease of implementation. Spherical microphone arrays can capture more sound field information than planar arrays. The collected multichannel speech signals can be transformed from the space domain to the spherical harmonic domain (SHD) for processing through spherical modal decomposition. The spherical harmonic domain MUSIC (SHD-MUSIC) algorithm reduces the dimensionality of the covariance matrix and achieves better DOA estimation performance than the conventional MUSIC algorithm. However, the SHD-MUSIC algorithm is prone to failure in low signal-to-noise ratio (SNR), high reverberation time (RT), and other multi-source environments. To address these challenges, we propose a novel joint spherical harmonic domain beam-space MUSIC (SHD-BMUSIC) algorithm in this paper. The advantage of decoupling the signal frequency and angle information in the SHD is exploited to improve the anti-reverberation property of the DOA estimation. In the SHD, the broadband beamforming matrix converts the SHD sound pressure to the beam domain output. Beamforming enhances the incoming signal in the desired direction and reduces the SNR threshold as well as the dimension of the signal covariance matrix. In addition, the 3D beam of the spherical array has rotational symmetry and its beam steering is decoupled from the beam shape. Therefore, the broadband beamforming constructed in this paper allows for the arbitrary adjustment of beam steering without the need to redesign the beam shape. Both simulation experiments and practical tests are conducted to verify that the proposed SHD-BMUSIC algorithm has a more robust adjacent source discrimination capability than the SHD-MUSIC algorithm.

On the application of MUSIC algorithm for identifying short sound-hard arcs in limited-view inverse acoustic problem

For proper application of the MUltiple SIgnal Classification (MUSIC) algorithm for imaging short, linear sound-hard arcs, one needs a priori information of the unit outward normal vector on each arc. Due to this reason, a set of directions was applied to estimate, which caused the imaging procedure’s slowness. Applying a unit vector instead of the estimation is natural to avoid this difficulty. However, no relevant mathematical theory has been developed to explain the various phenomena in simulation results. Here, we consider applying MUSIC without a priori information of the arcs. To this end, we introduce the imaging functions of MUSIC with and without information on the unit outward vector on the arcs. We analyze their mathematical structure by establishing an infinite series of Bessel functions and the range of incident/observation directions and discuss their properties. This is based on the physical factorization of the multi-static response (MSR) matrix that separates the incoming plane-wave information from the unknown density information. Numerical simulation results with noisy data are conducted for small and extended arcs to support the theoretical result.

Motor-Imagery EEG Signal Classification using Optimized Support Vector Machine by Differential Evolution Algorithm

Background: Motor-Imagery (MI) is a mental or cognitive stimulation without actual sensory input that enables the mind to represent perceptual information. This study aims to use the optimized support vector machine (OSVM) by differential evolution algorithm for motor-Imagery EEG signal classification. Methods: A total of three filters were applied to each signal during the preprocessing phase. The bandstop filter was used to remove urban noise and signal recorders, the median filter to remove random sudden peaks in the signal, and finally, the signal was normalized using the mapminmax filter. The most valuable features were extracted including mean signal intensity, minimum signal value, signal peak value, signal median, signal standard deviation, energy, corticoids, entropy, and signal skewness. Results: The accuracy of the SVM for linear, Gaussian, polynomial, and radial base kernels was 67.3%, 55.1%, 63.6%, and 55.1%, respectively, which was optimized after the classification model by differential evolution algorithm; however, the accuracy for OSVM was increased to 99.6%. Conclusion: Examination of the brain signal appearance for uniform motor-Imagery of both hands showed a significant difference between the signal of motor-Imagery mode with OSVM algorithm (99.6% accuracy), which gave promising results for classification motor imagery EEG signal.

Human Activity Tracker and Recognition

Human Activity Recognition (or, HAR) is a piece of software that uses AI algorithms to recognize and categories human physical activity. By analyzing signal data from multiple sensors such as accelerometers, gyroscopes, and magnetometers, the system is meant to recognize and categorize physical activities such as walking, running, leaping, ascending stairs, and others. To recognize human activity patterns, the HAR system employs signal preprocessing, feature extraction, and classification algorithms. The use of simulated intelligence techniques such as deep learning computations, convolutional brain organizations, and supporting vector machines has improved the display of HAR frameworks. The system may be utilized for a variety of purposes, including security, sports, fitness, and healthcare. In general, the HAR framework provides a beneficial value to robotized human activities. Man-made reasoning (Artificial Intelligence) plays an important role in Human Activity Recognition by allowing frameworks to learn and adapt to new conditions. In general, HAR framework is beneficial asset to robotized human movement recognition, working with the advancement of clever frameworks that can research human behaviour and work on personal fulfilment. Overall, Human Activity Recognition Using Computerized Reasoning is promising innovation that enables intelligent frameworks to perceive and group human activities gradually. This breakthrough has the potential to disrupt several businesses and improve people's personal pleasure by enabling personalized medical treatment, improving game execution, and improving street safety. The creation of this software sets the path for more study into themes such as the relationship between individual health status and physical activity. Overall, creating a fruitful Human Action Acknowledgement project utilizing recordings necessitates a broad understanding of AI and Profound Learning methods. As a result, success of this project highlights the value of creativity and perseverance in learning. Finally, it is the initial step towards developing more advanced systems that will improve people's lives in the future.

Soft Fall Detection With a Height-Tracking Method Based on MIMO Radar System

Radar-based fall detection technology has attracted much attention for its high accuracy, robustness, and privacy preservation potential. Detection of “Soft Fall”, i.e., high-freedom fall, is the key to practical application. This paper proposes a novel height-tracking method based on a multiple-input multiple-output (MIMO) radar system to address this problem. First, the received signal was segmented into a time sequence with a sliding window along slow time. Next, Fast Fourier Transform (FFT) and Multiple Signal Classification (MUSIC) algorithms were applied to estimate the general trend of the human body’s time-varying range and pitch angle information. Then, they were fused with a geometrical relationship to describe height changes during fall motions using a height trajectory map. Two height-based features were extracted as input to Support Vector Machine (SVM) to distinguish soft fall and fall-similar motions. Finally, experiments, including four soft fall and five typical fall-similar motions, were conducted to demonstrate its feasibility and superiority.

How to Determine an Optimal Noise Subspace?

The multiple signal classification (MUSIC) algorithm based on the orthogonality between the signal subspace and noise subspace is one of the most frequently used methods in the estimation of direction of arrival (DOA), and its performance of DOA estimation mainly depends on the accuracy of the noise subspace. In the most existing researches, the noise subspace is formed by (defined as) the eigenvectors corresponding to all small eigenvalues of the array output covariance matrix. However, we found that the estimation of DOA through the noise subspace in the traditional formation is not optimal in almost all cases, and using a partial noise subspace can always obtain optimal estimation results. In other words, the subspace spanned by the eigenvectors corresponding to a part of the small eigenvalues is more representative of the noise subspace. We demonstrate this conclusion through a number of experiments. Thus, it seems that which and how many eigenvectors should be selected to form the partial noise subspace would be an interesting issue. In addition, this research poses a much general problem: how to select eigenvectors to determine an optimal noise subspace?

A Time-Series Signal Classification Algorithm and Its Application to Nanopore Ionic Current Signal Identification★

A Time-Series Signal Classification Algorithm and Its Application to Nanopore Ionic Current Signal Identification^★

An Electrocardiogram Signal Classification Algorithm Based on Improved Deep Residual Shrinkage Networks

An Electrocardiogram Signal Classification Algorithm Based on Improved Deep Residual Shrinkage Networks