Search Results (83)

The accurate and early detection of respiratory diseases is vital for effective diagnosis and treatment. This study presents a new approach for classifying lung sounds using a double denoising method combined with a 1D Convolutional Neural Network (CNN). The preprocessing uses Fast Fourier Transform to clean up sounds and High-Pass Filtering to improve the quality of breathing sounds by eliminating noise and low-frequency interruptions. The Short-Time Fourier Transform (STFT) extracts features that capture localised frequency variations, crucial for distinguishing normal and abnormal respiratory sounds. These features are input into the 1D CNN, which classifies diseases such as bronchiectasis, pneumonia, asthma, COPD, healthy, and URTI. The dual denoising method enhances signal clarity and classification performance. The model achieved 96% validation accuracy, highlighting its reliability in detecting respiratory conditions. The results emphasise the effectiveness of combining signal augmentation with deep learning for automated respiratory sound analysis, with future research focusing on dataset expansion and model refinement for clinical use. Full article

This study demonstrates a rich complexity of the time–frequency ionospheric signal spectrum, dependent on the measurement type and platform. Different phenomena contributing to satellite-derived and ground-derived geophysical data that only selected signal bands can be potentially sensitive to seismicity over time, and they are applicable in lithosphere–atmosphere–ionosphere coupling (LAIC) studies. In this study, satellite-derived and ground-derived ionospheric observations are filtered by a Fourier-based band-pass filter, and an experimental selection of potentially sensitive frequency bands has been carried out. This work focuses on band-pass filtered ionospheric observations and seismic activity in the region of the Aegean Sea over a two-year time period (2020–2021), with particular focus on the entire system of tectonic plate junctions, which are suspected to be a potential source of ionospheric disturbances distributed over hundreds of kilometers. The temporal evolution of seismicity power in the Aegean region is represented by the record of earthquakes characterized by M ≥ 4.5, used for the estimation of cumulative seismic energy. The ionospheric response to LAIC is explored in three data types: short inspections of in situ electron density (Ne) over a tectonic plate boundary by Swarm satellites, stationary determination of three Ne density profile parameters by the Athens Digisonde station AT138 (maximum frequency of the F2 layer: foF2; maximum frequency of the sporadic E layer: foEs; and frequency spread: ff), and stationary measure of vertical total electron content (VTEC) interpolated from a UPC-IonSAT Quarter-of-an-hour time resolution Rapid Global ionospheric map (UQRG) near Athens. The spectrograms are made with the use of short-term Fourier transform (STFT). These frequency bands in the spectrograms, which show a notable coincidence with seismicity, are filtered out and compared to cumulative seismic energy in the Aegean Sea, to the geomagnetic Dst index, to sunspot number (SN), and to the solar radio flux (F10.7). In the case of Swarm, STFT allows for precise removal of long-wavelength Ne signals related to specific latitudes. The application of STFT to time series of ionospheric parameters from the Digisonde station and GIM VTEC is crucial in the removal of seasonal signals and strong diurnal and semi-diurnal signal components. The time series formed from experimentally selected wavebands of different ionospheric observations reveal a moderate but notable correlation with the seismic activity, higher than with any solar radiation parameter in 8 out of 12 cases. The correlation coefficient must be treated relatively and with caution here, as we have not determined the shift between seismic and ionospheric events, as this process requires more data. However, it can be observed from the spectrograms that some weak signals from selected frequencies are candidates to be related to seismic processes. Full article

As the application of the Global Navigation Satellite System (GNSS) continues to expand, its stability and safety issues are receiving more and more attention, especially the interference problem. Interference reduces the signal reception quality of ground terminals and may even lead to the paralysis of GNSS function in severe cases. In recent years, Low Earth Orbit (LEO) satellites have been highly emphasized for their unique advantages in GNSS interference detection, and related commercial and academic activities have increased rapidly. In this context, based on the signal-to-noise ratio (SNR) and radio-frequency interference (RFI) measurements data from COSMIC-2 satellites, this paper explores a method of predicting RFI measurements using SNR correlation variations in different GNSS signal channels for application to the detection and localization of civil terrestrial GNSS interference signals. Research shows that the SNR in different GNSS signal channels shows a correlated change under the influence of RFI. To this end, a CNN-BiLSTM-Attention model combining a convolutional neural network (CNN), bi-directional long and short-term memory network (BiLSTM), and attention mechanism is proposed in this paper, and the model takes the multi-channel SNR time series of the GNSS as the input and outputs the maximum measured value of RFI in the multi-channels. The experimental results show that compared with the traditional band-pass filtering inter-correlation method and other deep learning models, the model in this paper has a root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (R²) of 1.0185, 1.8567, and 0.9693, respectively, in RFI prediction, which demonstrates a higher RFI detection accuracy and a wide range of rough localization capabilities, showing significant competitiveness. Since the correlation changes in the SNR can be processed to decouple the signal strength, this model is also suitable for future GNSS-RO missions (such as COSMIC-1, CHAMP, GRACE, and Spire) for which no RFI measurements have yet been made. Full article

The scope of this work is to propose a distributed RF sensing architecture that interconnects and utilizes a cyber security operations center (SOC) to support long-term RF threat monitoring, alerting, and further centralized processing. For the purpose of this work, RF threats refer mainly to RF jamming, since this can jeopardize multiple wireless systems, either directly as a Denial of Service (DoS) attack, or as a means to force a cellular or WiFi wireless client to connect to a malicious system. Furthermore, the possibility of the suggested architecture to monitor signals from malicious drones in short distances is also examined. The work proposes, develops, and examines the performance of RF sensing sensors that can monitor any frequency band within the range of 1 MHz to 8 GHz, through selective band pass RF filtering, and subsequently these sensors are connected to a remote SOC. The proposed sensors incorporate an automatic calibration and time-depended environment RF profiling algorithm and procedure for optimizing RF jamming detection in a dense RF spectrum, occupied by heterogeneous RF technologies, thus minimizing false-positive alerts. The overall architecture supports TCP/IP interconnections of multiple RF jamming detection sensors through an efficient MQTT protocol, allowing the collaborative operation of sensors that are distributed in different areas of interest, depending on the scenario of interest, offering holistic monitoring by the centralized SOC. The incorporation of the centralized SOC in the overall architecture allows also the centralized application of machine learning algorithms on all the received data. Full article

This paper proposes a fractional order integral-derivative plus second-order derivative with low-pass filters and a tilt controller called I^λDND²N²-T to improve the control performance of an automatic voltage regulator (AVR). In this study, equilibrium optimisation (EO), multiverse optimisation (MVO), and particle swarm optimisation (PSO) algorithms are used to optimise the parameters of the proposed controller and statistical tests are performed with the data obtained from the application of these three algorithms to the AVR problem. Afterwards, the performance of the I^λDND²N²-T controller is demonstrated by comparing the transient responses with the results obtained in recently published papers. In addition, extra disturbances such as frequency deviation, load variation, and short circuit faults in the generator are applied to the AVR system. The proposed controller has outperformed the compared controller against these disturbances. Finally, a robustness test is performed in terms of deterioration in the system parameters. The results show that the I^λDND²N²-T controller outperforms the compared controllers under all conditions and exhibits a robust effect on the perturbed system parameters. Full article

Pipeline leakage represents a critical challenge in smart cities and various industries, leading to severe economic, environmental, and safety consequences. Early detection of leaks is essential for overcoming these risks and ensuring the safe operation of pipeline systems. In this study, a hybrid convolutional neural network–long short-term memory (CNN-LSTM) model for pipeline leak detection that uses acoustic emission signals was designed. In this model, acoustic emission signals are initially preprocessed using a Savitzky–Golay filter to reduce noise. The filtered signals are input into the hybrid model, where spatial features are extracted using a CNN. The features are then passed to an LSTM network, which extracts temporal features from the signals. Based on these features, the presence or absence of a leakage is determined. The performance of the proposed model was compared with two alternative approaches: a method that employs combined features from the time domain and LSTM and a bidirectional gated recurrent unit model. The proposed approach demonstrated superior performance, as evidenced by lower validation loss, higher validation accuracy, enhanced confusion matrices, and improved t-distributed stochastic neighbor embedding plots compared to the other models when tested on industrial data. The findings indicate that the proposed model is more effective in accurately detecting pipeline leaks, offering a promising solution for enhancing smart cities and industrial safety. Full article

To address the performance and computational complexity issues in speech enhancement for hearing aids, a speech enhancement algorithm based on a microphone array and a lightweight two-stage convolutional recurrent network (CRN) is proposed. The algorithm consists of two main modules: a beamforming module and a post-filtering module. The beamforming module utilizes directional features and a complex time-frequency long short-term memory (CFT-LSTM) network to extract local representations and perform spatial filtering. The post-filtering module uses analogous encoding and two symmetric decoding structures, with stacked CFT-LSTM blocks in between. It further reduces residual noise and improves filtering performance by passing spatial information through an inter-channel masking module. Experimental results show that this algorithm outperforms existing methods on the generated hearing aid dataset and the CHIME-3 dataset, with fewer parameters and lower model complexity, making it suitable for hearing aid scenarios with limited computational resources. Full article

Bearings are the most vulnerable component in low-voltage induction motors from a maintenance standpoint. Vibration monitoring is the benchmark technique for identifying mechanical faults in rotating machinery, including the diagnosis of bearing defects. The study of different bearing fault phenomena under induction motor transient conditions offers interesting capabilities to enhance classic fault detection techniques. This study analyzes the low-frequency localized bearing fault signatures in both the inner and outer races during the start-up and steady-state operation of inverter-fed and line-started induction motors. For this aim, the classic vibration envelope spectrum technique is explored in the time–frequency domain by using a simple, resampling-free, Short Time Fourier Transform (STFT) and a band-pass filtering stage. The vibration data are acquired in the motor housing in the radial direction for different load points. In addition, two different localized defect sizes are considered to explore the influence of the defect width. The analysis of extracted low-frequency characteristic frequencies conducted in this study demonstrates the feasibility of detecting early-stage localized bearing defects in induction motors across various operating conditions and actuation modes. Full article

This paper describes a revolutionary design paradigm for monitoring aquatic life. This unique methodology addresses issues such as limited memory, insufficient bandwidth, and excessive noise levels by combining two approaches to create a comprehensive predictive filtration system, as well as multiple-transfer route analysis. This work focuses on proposing a novel filtration learning approach for underwater sensor nodes. This model was created by merging two adaptive filters, the finite impulse response (FIR) and the adaptive line enhancer (ALE). The FIR integrated filter eliminates unwanted noise from the signal by obtaining a linear response phase and passes the signal without distortion. The goal of the ALE filter is to properly separate the noise signal from the measured signal, resulting in the signal of interest. The cluster head level filters are the adaptive cuckoo filter (ACF) and the Kalman filter. The ACF assesses whether an emitter node is part of a set or not. The Kalman filter improves the estimation of state values for a dynamic underwater sensor networking system. It uses distributed learning long short-term memory (LSTM-CNN) technology to ensure that the anticipated value of the square of the gap between the prediction and the correct state is the smallest possible. Compared to prior methods, our suggested deep filtering–learning model achieved 98.5% of the sensory filtration method in the majority of the obtained data and close to 99.1% of an adaptive prediction method, while also consuming little energy during lengthy monitoring. Full article

Alpha emitters like plutonium pose severe health risks when ingested, damaging DNA and potentially causing cancer. Traditional detection methods require proximity within millimeters of the contamination source, presenting safety risks and operational inefficiencies. Long-range detection through alpha radioluminescence (RL) offers a promising alternative. However, most of the previous experiments have been carried out under controlled conditions that preclude the overwhelming effect of ambient light. This study demonstrates the successful detection of a 3 MBq alpha emitter in an open environment using a compact alpha camera. This camera incorporates a deep-cooled CCD and a low f-number lens system designed to minimize the blue shift effects of filters. Night-time imaging was achieved with a dual-filter system using a sandwich filter assembly centered at 337 nm and 343 nm for capturing alpha RL and subtracting background light, respectively. At night, the alpha source was detected from 1 m away within one minute, and the lowest detection limit can be calculated as 75 kBq. The system was also evaluated under simulated urban lighting conditions. For daytime imaging, a stack of tilted 276 nm short pass filters minimized sunlight interference, enabling the detection of the alpha source at 70 cm within 10 min under indirect sunlight. This research highlights the viability of long-range optical detection of alpha emitters for environmental monitoring in real-world settings. Full article

Brain-inspired flexible neuromorphic devices are of great significance for next-generation high-efficiency wearable sensing and computing systems. In this paper, we propose a flexible organic electrochemical transistor using poly[(bithiophene)-alternate-(2,5-di(2-octyldodecyl)- 3,6-di(thienyl)-pyrrolyl pyrrolidone)] (DPPT-TT) as the organic semiconductor and poly(methyl methacrylate) (PMMA)/LiClO₄ solid-state electrolyte as the gate dielectric layer. Under gate voltage modulation, an electric double layer (EDL) forms between the dielectric layer and the channel, allowing the device to operate at low voltages. Furthermore, by leveraging the double layer effect and electrochemical doping within the device, we successfully mimic various synaptic behaviors, including excitatory post-synaptic currents (EPSC), paired-pulse facilitation (PPF), high-pass filtering characteristics, transitions from short-term plasticity (STP) to long-term plasticity (LTP), and demonstrate its image recognition and storage capabilities in a 3 × 3 array. Importantly, the device’s electrical performance remains stable even after bending, achieving ultra-low-power consumption of 2.08 fJ per synaptic event at −0.001 V. This research may contribute to the development of ultra-low-power neuromorphic computing, biomimetic robotics, and artificial intelligence. Full article

This paper proposes a deep learning (DL)-assisted high-pass-filter (HPF)-based fixed-threshold decision (FTD) for free-space optical (FSO) communication. HPF is applied to reduce the scintillation effect by filtering out the low-frequency components of the received signal. However, the performance is limited owing to the signal distortion from HPF and remnant scintillation effect due to insufficient filtering. Therefore, the DL model is adopted to improve the performance of HPF-based scintillation effect compensation. The multilayer perceptron (MLP) model is used to adaptively select the peak frequency component of the received signal as the optimized cutoff frequency of HPF. Furthermore, recurrent neural network (RNN) and long short-term memory (LSTM) models are cascaded after HPF to compensate for the remnant scintillation effect and recover the signal distortion without the optimization of HPF cutoff frequency. The simulation was conducted under different turbulence channels and data rates. Simulation results showed that MLP-assisted adaptive optimized cutoff frequency and cascaded LSTM and HPF methods were close to the adaptive-threshold decision with precise channel state information under various turbulence channel degrees. Full article

The high-frequency (HF) voltage injection method is widely applied in achieving position-sensorless control for interior permanent-magnet synchronous motors (IPMSMs). This method necessitates precise and rapid extraction of the current signal for accurate position estimation and field-oriented control (FOC). In the traditional methods, the position error signal and fundamental current are extracted from the current signal using band-pass filters (BPFs) and low-pass filters (LPFs), or a method based on time-delay filters. However, the traditional extraction method falls short in ensuring simultaneous dynamic performance and accuracy, particularly when the switching frequency is limited or when encountering harmonic and noise interference. In this article, a novel HF pulsating square-wave voltage injection method based on an improved current signal-extraction strategy is proposed to improve the extraction accuracy while maintaining good dynamic performance. The newly devised current signal-extraction method is crafted upon a notch filter (NF). Through harnessing NF’s effective separation characteristics of specific frequency signals, the current signal is meticulously processed. This process yields the extraction of the position error signal and fundamental-current component, crucial for accurate position estimation and motor FOC. Simulation and hardware-in-the-loop (HIL) testing are conducted to validate the effectiveness of the proposed approach. Full article

The problem of energy depletion has brought wind energy under consideration to replace oil- or chemical-based energy. However, the breakdown of wind turbines is a major concern. Accordingly, unsupervised learning was performed using the vibration signal of a wind power generator to achieve an outlier detection performance of 97%. We analyzed the vibration data through wavelet packet conversion and identified a specific frequency band that showed a large difference between the normal and abnormal data. To emphasize these specific frequency bands, high-pass filters were applied to maximize the difference. Subsequently, the dimensions of the data were reduced through principal component analysis, giving unique characteristics to the data preprocessing process. Normal data collected from a wind farm located in northern Sweden was first preprocessed and trained using a long short-term memory (LSTM) autoencoder to perform outlier detection. The LSTM Autoencoder is a model specialized for time-series data that learns the patterns of normal data and detects other data as outliers. Therefore, we propose a method for outlier detection through data preprocessing and unsupervised learning, utilizing the vibration signals from wind generators. This will facilitate the quick and accurate detection of wind power generator failures and provide alternatives to the problem of energy depletion. Full article

Short-range Internet of Things (IoT) sensor nodes operating at 2.4 GHz must provide ubiquitous wireless sensor networks (WSNs) with energy-efficient, wide-range output power (POUT). They must also be fully integrated on a single chip for wireless body area networks (WBANs) and wireless personal area networks (WPANs) using low-power Bluetooth (BLE) and Zigbee standards. The proposed fully integrated transmitter (TX) utilizes a digitally controllable current-mode class-D (CMCD) power amplifier (PA) with a second harmonic distortion (HD2) suppression to reduce VCO pulling in an integrated system while meeting harmonic limit regulations. The CMCD PA is divided into 7-bit slices that can be reconfigured between differential and single-ended topologies. Duty cycle distortion compensation is performed for HD2 suppression, and an HD2 rejection filter and a modified C-L-C low-pass filter (LPF) reduce HD2 further. Implemented in a 28 nm CMOS process, the TX achieves a wide P_OUT range of from 12.1 to −31 dBm and provides a maximum efficiency of 39.8% while consuming 41.1 mW at 12.1 dBm P_OUT. The calibrated HD2 level is −82.2 dBc at 9.93 dBm P_OUT, resulting in a transmitter figure of merit (TX_FoM) of −97.52 dB. Higher-order harmonic levels remain below −41.2 dBm even at 12.1 dBm P_OUT, meeting regulatory requirements. Full article