Frequency Domain Data Research Articles

Visualization of Acoustic Properties of Thin Films Based on Frequency Domain Data

Visualization of Acoustic Properties of Thin Films Based on Frequency Domain Data

Frequency domain LPV controller synthesis for a positioning system with uncertain scheduling parameters

Ever-increasing demands from the industry require better performance out of the same system to achieve a competitive advantage. In systems functioning at different operating points, a fundamental trade-off between robustness and performance can limit the achieved performance. Linear Parameter Varying (LPV) controller synthesis can overcome this issue by adapting the controller parameters with the operating points. This requires a parametric LPV model of the system, which in many cases is difficult or costly to obtain. Recently, LPV controller design methods based on the frequency-domain data in different operating points have shown to result in great performance without the need of a parametric LPV model. In these approaches, however, the scheduling parameters are assumed to be measured exactly. In this paper, a new approach is proposed to design LPV controllers with slowly varying uncertain scheduling parameters, using only the frequency response data of a SISO system around different operating points. The proposed approach is applied on a rotary table, a machine used in the manufacturing industry. The dynamics of the rotary table depend on the inertia of the object mounted on-top which cannot be directly measured. Since the scheduling parameter cannot be measured, it is estimated using a short sequence of data that leads to an inexact estimation with interval uncertainty. For the system considered, the LPV controller designed by the proposed approach shows clear improvement over the state-of-the-art robust controller synthesis approaches.

Cardiovascular autonomic responses in reproductive and post-menopausal women

Background: Autonomic responses are a sensitive indicator for assessing balance in sympatho-vagal system. Many studies have shown an increased response in the sympathetic system in post-menopausal women. Alteration in autonomic nerve function causes an increased incidence of hypertension, arrhythmias, or acute coronary syndrome. Thus, this study aims to find out the heart rate variability (HRV) among reproductive and post-menopausal women. Aims and Objectives: The aim of the study was to compare the cardiovascular autonomic responses among reproductive and post-menopausal women. Materials and Methods: This cross-sectional study was conducted in the department of physiology at government medical college, Kottayam on 40 reproductive and 40 post-menopausal women. Using PHYSIOPAC-PP4 (Medicaid Systems Chandigarh) machine, an electrocardiogram was recorded in both groups in the supine position for 5 min. The frequency and time domain data obtained were analyzed using SPSS software. Results: The mean age group in reproductive women was 43 and post-menopausal women was 54. In the power spectrum analysis of HRV parameters the mean value of frequency domain Low frequency (LF) in reproductive women was 72.05 (S.D = 5.706) and post-menopausal women was 79.48 (S.D = 3.789). The high frequency (HF) in reproductive women was 28 (S.D = 5.588) and post-menopausal women were 20.53 (S.D = 3.789). The LF/HF ratio mean value obtained in reproductive women was 2.723 (S. D = 0.759) and post–menopausal was 3.992 (S.D 0.785). In the various time domain parameters, we estimated the mean value of the root mean square of successive differences (RMSSD) between normal heartbeats RMSSD and obtained the value in reproductive women as 25.05 (S.D = 11.88) and post-menopausal women as 12.68 (S.D = 5.535). The RR interval triangular index mean value in reproductive women was 0.0520 (S.D = 0.013) and post-menopausal women were 0.0345 (S.D = 0.01). In all these parameters, the P-value was found to be <0.001, so statistically highly significant. Conclusion: In this study, higher values in time domain and frequency domain parameters for assessing sympathetic functions were seen in postmenopausal women when compared to women of the reproductive age group which was statistically significant. This may be due to age and hormonal influences. The power spectrum analysis of HRV parameters was proved to be a noninvasive method to assess the sympathovagal balance in the study group of women.

Frequency-Domain Electromagnetic and VLF Data with An Application to Modeling of Basement Structures: A Case Study Within, Ibadan Area, Southwestern Nigeria

The study area, the church camp is within Ibadan which fall within southwestern basement complex of Nigeria and it is mainly by migmatite-gneiss. The hard rocks are usually characterized by basement fracture which may serve as a conduit for groundwater passage. This has necessitated detail geological and geophysical investigation to accurately and precisely delineate this structure. Geological mapping alongside very low frequency Electromagnetic (VLF-EM) techniques were adopted in which conductivity data were acquired along Fifteen (15) VLF profiles using ABEM WADI. Thirteen (13) of these profiles were in the East-West (E-W) direction which is the dip direction of the structural elements, while the remaining two profiles were in North-South (N-S) direction. The data obtained from the field were processed, filtered and presented inform of profiles and Karous-Hjelt (K-H) filtered pseudo-section to visualise conductivity in (2-D). Pockets of conductive structures were delineated and were interpreted as fractures of various dimension and orientation. Some are single fracture and others are closely-spaced double fractures which some of them joined or fused together at depth with some suspected to contain water while some are dry base on their conductivity/resistivity signature exhibited. In conclusion, VLF-EM has proven to possess the capability to characterize and model the basement fractures and define their dimension, axial orientation and indicates their parallelism which also signifies their origin as regard stress regime that produces the fracture.

Interpretation of Frequency-Domain Airborne Electromagnetic Data Based on the Deep Neural Network Incorporating Topographic Information

Deep neural networks (DNNs) have recently been used to interpret frequency-domain electromagnetic (EM) data, therefore vast amounts of information can be rapidly interpreted. However, in airborne surveys that include mountainous regions, severe topographic changes distort the EM data and they can lead to unreasonable interpretations. DNN-based inversions that reduce the effects of topographic changes have not yet been proposed whereas some conventional inversion techniques can correct topographic distortions. Since DNN-based interpretation needs various training dataset and its performance depends on the characteristics of training dataset, it is important to generate various AEM data and resistivity model pairs. However, it is almost impossible to derive the EM responses of two-dimensional (2D) or 3D resistivity models incorporating diverse topographic patterns because such models require many variables and computation costs are limited. Therefore, we suggested the pseudo 1D interpretation of EM data which can consider the topographic changes using EM responses and topographic information at three neighboring data points. We include topographic information in DNN training as slope angle between the data points. Compared to conventional inversion, our trained DNN model recovers the synthetic resistivity models more reasonably. In addition, we used the trained DNN model to evaluate a real AEM dataset from south-central Alaska. The trained DNN provided reasonable interpretation results despite that the data were acquired at mountainous area with rough topography. Since the trained DNN model can provide predictions rapidly, the interpretation time was drastically reduced. We believe that this research can increase the practical field applicability of AEM survey.

Multimodal Hierarchical CNN Feature Fusion for Stress Detection

Stress is one of the most severe concerns in modern life. High-level stress can create various diseases or loss of focus and productivity at work. Being under stress prevents people from recognizing their stress levels, so early stress detection is essential. Recently, multimodal fusion has enhanced the performance of stress detection models using Deep Learning (DL) techniques. The low, mid, and high-level features of a Convolutional Neural Network (CNN) are discriminative. A comprehensive feature representation can be obtained by fusing all three levels of CNN’s features. This study mainly focuses on detecting stress by exploiting these advantages using a multimodal hierarchical CNN feature fusion. The two multimodal physiological signals used in this study are Electrodermal activity (EDA) and Electrocardiogram (ECG). We develop a hierarchical feature set by concatenating multi-level CNN features for each modality. Multimodal fusion on both hierarchical feature sets is performed using the Multimodal Transfer Module (MMTM). The experiments are carried out with raw frequency domain data and the features from the frequency bands to study the effectiveness of both. The model’s performance is compared to the different combinations of hierarchical features from low, mid, and high levels. To verify the generalizability, the proposed approach has been evaluated on four benchmark datasets - ASCERTAIN, CLAS, MAUS, and WAUC. The proposed method showed its effectiveness by outperforming existing models by 1-2%, respectively, on frequency band features. It is observed that the hierarchical feature set from all three levels performed better than all other combinations by 2-4%. As a result, this strategy can be a useful addition to stress detection.

Cloud-enabled Internet of Things Medical Image Processing Compressed Sensing Reconstruction

Deep learning compresses medical image processing in IoMT. CS-MRI acquires quickly. It has various medicinal uses due to its advantages. This lowers motion artifacts and contrast washout. Reduces patient pressure and scanning costs. CS-MRI avoids the Nyquist-Shannon sampling barrier. Parallel imagingbased fast MRI uses many coils to reconstruct MRI images with less raw data. Parallel imaging enables rapid MRI. This research developed a deep learning-based method for reconstructing CS-MRI images that bridges the gap between typical non-learning algorithms that employ data from a single image and enormous training datasets. Conventional approaches only reconstruct CS-MRI data from one picture. Reconstructing CS-MRI images. CS-GAN is recommended for CS-MRI reconstruction. For success. Refinement learning stabilizes our C-GAN-based generator, which eliminates aliasing artifacts. This improved newly produced data. Product quality increased. Adversarial and information loss recreated the picture. We should protect the image’s texture and edges. Picture and frequency domain data establish consistency. We want frequency and picture domain information to match. It offers visual domain data. Traditional CS-MRI reconstruction and deep learning were used in our broad comparison research. C-GAN enhances reconstruction while conserving perceptual visual information. MRI image reconstruction takes 5 milliseconds, allowing real-time analysis.

A Data Enhancement Method toward Real-World Single Image Super-Resolution

Introducing convolutional neural networks (CNNs) into single image super-resolution (SISR) has achieved remarkable performance in experimental scenarios over the past decades. However, maintaining constant mapping complexity, most existing CNN-based SISR approaches make it challenging to reconstruct high-quality images in real scenarios because of considerable variation between authentic degradation and simulated degradation. In this paper, to improve SISR models’ capabilities in the real world, we propose a data enhancement method, consisting of frequency domain data augmentation (FDA) and dynamical frequency loss (DFL). By introducing high-resolution images’ frequency information, the FDA enhances the quality and quantity of data, regularizing the model to learn “how” to appropriately apply super-resolution on different frequency components of a given image. In addition, to further strengthen the regularization effect, we design DFL to dynamically assign the loss trade-off hyperparameters of distinct frequency components, guiding the model to allocate computational overhead reasonably. Considering quantitative analysis and visual quality in combination, extensive experiments prove that the proposed enhancement method significantly improves SR models’ capabilities in real scenarios without adding any computational cost to the evaluation procedure.

Time Domain Near-Infrared Optical Tomography Utilizing Full Temporal Data: A Simulation Study.

The analysis of full temporal data in time-domain near-infrared optical tomography (TD NIROT) measurements enables valuable information to be obtained about tissue properties with good temporal and spatial resolution. However, the large amount of data obtained is not easy to handle in the image reconstruction. The goal of the project is to employ full-temporal data from a TD NIROT modality. We improved TD data-based 3D image reconstruction and compared the performance with other methods using frequency domain (FD) and temporal moments. The iterative reconstruction algorithm was evaluated in simulations with both noiseless and noisy in-silico data. In the noiseless cases, a superior image quality was achieved by the reconstruction using full temporal data, especially when dealing with inclusions at 20 mm and deeper in the tissue. When noise similar to measured data was present, the quality of the recovered image from full temporal data was no longer superior to the one obtained from the analysis of FD data and temporal moments. This indicates that denoising methods for TD data should be developed. In conclusion, TD data contain richer information and yield better image quality.

Unsupervised multimodal domain adversarial network for time series classification

Unsupervised Domain Adaptation (UDA) is an ideal transfer learning method, which can use labeled source data to improve the classification performance of unlabeled target data. At present, the UDA methods for Time Series Classification (TSC) only use time-domain data or frequency-domain data as the input, and ignore fusing them, resulting in insufficient feature extraction and inaccurate source-target distribution alignment. Therefore, we propose an unsupervised Multimodal Domain Adversarial Network (MDAN) for TSC tasks. Specifically, we adopt two feature extractors for the time-domain and frequency-domain feature representations, and employ-three classifiers to perform TSC of source data for training the two feature extractors; Then, we fuse the time-domain and frequency-domain feature representations of source and target data, respectively, input them into the unified domain discriminator for unsupervised multimodal domain adversarial learning, and combine the proposed Time-Frequency-domain Joint Maximum Mean Discrepancy (TF-JMMD) to accurately align the source-target distributions; Finally, we select CNN or ResNet18 as the feature extractors to carry out comprehensive experiments, and the results demonstrate the SOTA performance of MDAN.

Robust Wavelet Transform Neural-Network-Based Short-Term Load Forecasting for Power Distribution Networks

A precise short-term load-forecasting model is vital for energy companies to create accurate supply plans to reduce carbon dioxide production, causing our lives to be more environmentally friendly. A variety of high-voltage-level load-forecasting approaches, such as linear regression (LR), autoregressive integrated moving average (ARIMA), and artificial neural network (ANN) models, have been proposed in recent decades. However, unlike load forecasting in high-voltage transmission systems, load forecasting at the distribution network level is more challenging since distribution networks are more variable and nonstationary. Moreover, existing load-forecasting models only consider the features of the time domain, while the demand load is highly correlated to the frequency-domain information. This paper introduces a robust wavelet transform neural network load-forecasting model. The proposed model utilizes both time- and frequency-domain information to improve the model’s prediction accuracy. Firstly, three wavelet transform methods, variational mode decomposition (VMD), empirical mode decomposition (EMD), and empirical wavelet transformation (EWT), were introduced to transform the time-domain demand load data into frequency-domain data. Then, neural network models were trained to predict all components simultaneously. Finally, all the predicted data were aggregated to form the predicted demand load. Three cases were simulated in the case study stage to evaluate the prediction accuracy under different layer numbers, weather information, and neural network types. The simulation results showed that the proposed robust time–frequency load-forecasting model performed better than the traditional time-domain forecasting models based on the comparison of the performance metrics, including the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean squared error (RMSE).

DRG-Keyboard

We present DRG-Keyboard, a gesture keyboard enabled by dual IMU rings, allowing the user to swipe the thumb on the index fingertip to perform word gesture typing as if typing on a miniature QWERTY keyboard. With dual IMUs attached to the user's thumb and index finger, DRG-Keyboard can 1) measure the relative attitude while mapping it to the 2D fingertip coordinates and 2) detect the thumb's touch-down and touch-up events combining the relative attitude data and the synchronous frequency domain data, based on which a fingertip gesture keyboard can be implemented. To understand users typing behavior on the index fingertip with DRG-Keyboard, we collected and analyzed user data in two typing manners. Based on the statistics of the gesture data, we enhanced the elastic matching algorithm with rigid pruning and distance measurement transform. The user study showed DRG-Keyboard achieved an input speed of 12.9 WPM (68.3% of their gesture typing speed on the smartphone) for all participants. The appending study also demonstrated the superiority of DRG-Keyboard for better form factors and wider usage scenarios. To sum up, DRG-Keyboard not only achieves good text entry speed merely on a tiny fingertip input surface, but is also well accepted by the participants for the input subtleness, accuracy, good haptic feedback, and availability.

Cutting Parameters and Material Classification Using Multinomial Logistic Regression

In the context of the new industrial revolution - Industry 4.0, the smart factory concept brought, to manufacturing, the idea of using large amounts of data acquired from a machining process and a set of mathematical techniques, discovering correlations, patterns, or trends in this database. Thus, machine tools are the focus of research in order to monitor and analyze the quality of the machining process based on data from embedded sensors. Based on this strategy, a signature process was created, that consists in capturing behavior patterns of a machine, such as machining conditions, machining quality, or tool wear. This article deal with a comparison between three Multinomial Logistic Regressions: the first using only time domain data, the second using only frequency domain data, and finally, the third using time and frequency domain data to identify the pattern of feed rate, depth of cut, and material being machined. It was observed that the methods had a precision of 96.25%, 37.92%, and 99.58%, respectively, showing that this methodology has great predictive efficiency and could be used to monitor the cutting parameters and material studied in this paper.

Field measurement and simplified numerical model for vibration response of subway superstructure

The residents near subway lines may be influenced by subway-induced vibration to a large extent. A simplified numerical modelling technique of the subway superstructure with a high-precision Caughey damping was proposed based on finite element software ANSYS in this paper, aiming to simulate the above-ground vibration response related to residents’ comfort under subway operation. Firstly, the two-stage field measurement of “free field-floor slabs of superstructure”, spanning four years, was carried out on a frame structure building above Guangzhou Metro Line 3. The time domain and frequency domain data of the measured vibration were analysed. A simplified numerical finite element (FE) model of the superstructure was then established, in which a high-precision Caughey damping was constructed using seven control frequencies. The comparative analysis of the simulated and measured results of the building verified that the simplified FE model for simulating the subway superstructure’s vibration response worked well and facilitated engineering applications. Additionally, another superstructure model of the shear wall was established by the same method to explore further the difference in the propagation characteristics of subway-induced vibration in the structure of two different forms. Finally, design recommendations for vibration control of subway superstructure were provided.

A Deep-Learning-Based Multi-Modal Sensor Fusion Approach for Detection of Equipment Faults

Condition monitoring is a part of the predictive maintenance approach applied to detect and prevent unexpected equipment failures by monitoring machine conditions. Early detection of equipment failures in industrial systems can greatly reduce scrap and financial losses. Developed sensor data acquisition technologies allow for digitally generating and storing many types of sensor data. Data-driven computational models allow the extraction of information about the machine’s state from acquired sensor data. The outstanding generalization capabilities of deep learning models have enabled them to play a significant role as a data-driven computational fault model in equipment condition monitoring. A challenge of fault detection applications is that single-sensor data can be insufficient in performance to detect equipment anomalies. Furthermore, data in different domains can reveal more prominent features depending on the fault type, but may not always be obvious. To address this issue, this paper proposes a multi-modal sensor fusion-based deep learning model to detect equipment faults by fusing information not only from different sensors but also from different signal domains. The effectiveness of the model’s fault detection capability is shown by utilizing the most commonly encountered equipment types in the industry, such as electric motors. Two different sensor types’ raw time domain and frequency domain data are utilized. The raw data from the vibration and current sensors are transformed into time-frequency images using short-time Fourier transform (STFT). Then, time-frequency images and raw time series data were supplied to the designed deep learning model to detect failures. The results showed that the fusion of multi-modal sensor data using the proposed model can be advantageous in equipment fault detection.

Adaptation of Range-Doppler Algorithm for Efficient Beamforming of Monostatic and Multistatic Ultrasound Signals.

Algorithmic changes that increase beamforming speed have become increasingly relevant to processing synthetic aperture (SA) ultrasound data. In particular, beamforming SA data in a spatio-temporal frequency domain using the F-k (Stolt) migration have been shown to reduce the beamforming time by up to two orders of magnitude compared with the conventional delay-and-sum (DAS) beamforming, and it has been used in applications where large amounts of raw data make real-time frame rates difficult to attain, such as multistatic SA imaging and plane-wave Doppler imaging with large ensemble lengths. However, beamforming signals in a spatio-temporal Fourier space can require loading large blocks of data at once, making it memory-intensive and less suited for parallel (i.e., multithreaded) processing. As an alternative, we propose beamforming in a range-Doppler (RD) frequency domain using the range-Doppler algorithm (RDA) that has originally been developed for SA radar (SAR) imaging. Through simulation and phantom experiments, we show that RDA achieves similar lateral resolution and contrast compared with DAS and F-k migration. At the same time, higher axial sidelobes in RDA images can be reduced via (temporal) frequency binning. Like the F-k migration, RDA significantly reduces the overall number of computations relative to DAS, and it achieves ten times lower processing time on a single CPU. Because RDA uses only a spatial Fourier transform (FT), it requires two times less memory than the F-k migration to process the simulated multistatic data and can be executed on as many as a thousand parallel threads (compared with eight parallel threads for the F-k migration), making it more suitable for implementation on modern graphics processing units (GPUs). While RDA is not as parallelizable as DAS, it is expected to hold a significant speed advantage on devices with moderate parallel processing capabilities (up to several thousand cores), such as point-of-care and low-cost ultrasound devices.

Spurious Modes in Model Order Reduction in Variational Problems in Electromagnetics

In this work, we address an everlasting issue in model order reduction (MOR) in electromagnetics that has remained unnoticed until now. Contrary to what has been previously done, we identify for the very first time spurious modes in MOR for time-harmonic Maxwell’s equations and propose a methodology to remove their negative influence on the reduced order model (ROM) response. These spurious modes have nonzero resonance frequency and may have shown up in the past giving rise to spikes in the frequency response, in effect, deteriorating the accuracy and efficiency of the MOR process. However, they were never characterized as spurious mode contributions, rather they were most likely considered as poor localized approximation issues in the MOR process. When we try to get further physical insights from the ROM, rather than simple frequency domain data, we cannot afford any poorly localized approximation issue, that is, any spurious mode in the band of analysis. Otherwise, these mathematical, but nonphysical, modes will mislead the physical behavior of the device under analysis. A computationally inexpensive variational divergence condition is established to identify spurious modes in the band of analysis, since any physical in-band mode must be divergence-free. In addition, once a spurious mode is identified in the band of analysis, its influence is removed from the ROM by a physics-based coupling strategy. As a result, a robust spurious mode contribution-free MOR in electromagnetics is proposed. Finally, several actual microwave circuits, such as a quad-mode filter and a triple-mode triple-band filter, will illustrate the capabilities and efficiency of the proposed approach.

Frequency domain data encoding in apache IoTDB

Frequency domain analysis is widely conducted on time series. While online transforming from time domain to frequency domain is costly, e.g., by Fast Fourier Transform (FFT), it is highly demanded to store the frequency domain data for reuse. However, frequency domain data encoding for efficient storage is surprisingly untouched. We notice that (1) the precision of data value is unnecessarily high after transforming to frequency domain and (2) the data values are with skewed distribution leading to a very large bit width for encoding. To avoid such space waste in both precision and skewness, we devise a descending bit-packing encoding for frequency domain data. Specifically, we quantize the data values in proper precision referring to the signal-noise-ratio (SNR) in frequency domain analysis. Moreover, we sort the data values in descending order so that the bit width could be dynamically reduced in encoding. The method has been deployed in Apache IoTDB, an open-source time-series database, not only for directly encoding frequency domain data, but also as a lossy compression of the time domain data. The extensive experiments on the system demonstrate the superiority of our encoding for both frequency domain and time domain data.

Dynamic load identification based on deep convolution neural network

The deep learning methods have been extensively studied in the field of dynamic load identification, due to their strong direct modeling ability between vibration response and external excitation. Dynamic load identification is a complicated inverse problem, which extremely relies on the solution of the structural model parameter. Nevertheless, the accurate computation of model parameters is always a challenge, small errors in model parameters will lead to inaccuracy of dynamic load identification. This brings various hardships to engineering applications. To achieve this problem, we propose a novel method based on a deep dilated convolution neural network (DCNN) for dynamic load identification, directly constructing the inverse model between vibration response and excitation, avoiding solving the model parameter. A dynamic load identification model, which contains two 1-D dilated convolution layers, one pooling layer, and two fully connected layers, is constructed to estimate the sinusoidal, impact, and random dynamic loads of a simply supported beam. We also appraise the anti-noise ability of the proposed method for load identification. Moreover, a vibration test is carried out to further evaluate this algorithm in the experimental aspect. Additionally, we analyze the comparison of the proposed method along with the Green kernel function method, and the dynamic system with uncertain model parameters is also analyzed. Besides, the operation of the convolution layer for response input is studied, and the applicability to different distributions of measurement points and the adaptability for frequency domain data are investigated. Ultimately, we find this method has a strong anti-noise ability due to its convolution layer, which can be regarded as a filter in dynamic load identification. Furthermore, the proposed method is of great practical for engineering applications owing to its satisfying applicability for systems with uncertain parameters, distributions of measurement points, and frequency data. All the results reveal the advantages of the identification method based on DCNN, consisting of good accuracy, reliability, and robustness. These results can be favorable for many applications.

JR-TFViT: A Lightweight Efficient Radar Jamming Recognition Network Based on Global Representation of the Time–Frequency Domain

Efficient jamming recognition capability is a prerequisite for radar anti-jamming and can enhance the survivability of radar in electronic warfare. Traditional recognition methods based on manually designed feature parameters have found it difficult to cope with the increasingly complex electromagnetic environment, and research combining deep learning to achieve jamming recognition is gradually increasing. However, existing research on radar jamming recognition based on deep learning can ignore the global representation in the jamming time–frequency domain data, while not paying enough attention to the problem of lightweighting the recognition network itself. Therefore, this paper proposes a lightweight jamming recognition network (JR-TFViT) that can fuse the global representation of jamming time–frequency domain data while combining the advantages of the Vision Transformer and a convolutional neural network (CNN). The global representation and local information in the jamming time–frequency data are fused with the assistance of the multi-head self-attention (MSA) mechanism in the transformer to improve the feature extraction capabilities of the recognition network. The model’s parameters are further decreased by modifying the standard convolutional operation mechanism and substituting the convolutional operation needed by the network with Ghost convolution, which has less parameters. The experimental results show that the JR-TFViT requires fewer model parameters while maintaining higher recognition performance than mainstream convolutional neural networks and lightweight CNNs. For 12 types of radar jamming, the JR-TFViT achieves 99.5% recognition accuracy at JNR = −6 dB with only 3.66 M model parameters. In addition, 98.9% recognition accuracy is maintained when the JR-TFViT parameter number is further compressed to 0.67 M.