Search

High power consumption is one of the main problems in the practical application of massive multi-user multi-input-multi-output (MU-MIMO) systems, and using 1-bit digital-to-analog converters (DACs) instead of high-resolution DACs is an effective way to reduce the power consumption of massive MU-MIMO systems. How to design efficient precoding schemes to improve the performance of 1-bit MU-MIMO systems has become a hot research field. This paper mainly studies the precoding problem of 1-bit DACs based on minimum mean squared error (MMSE). We first transform the non-convex constraints of the traditional optimization problem using binary equilibrium constraints. Then, to solve the transformed objective problem, we establish a framework based on the alternating direction method (ADM), use the fast projection gradient (FPG) method, and propose an alternating maximum minimum (AMM) algorithm to solve the optimization variables alternately. We have verified the convergence of the proposed algorithm and analyzed the accuracy of the proposed ADM framework and the computational complexity of the proposed algorithm in detail. In the simulation section, we analyze the uncoded bit error rate (BER) performance of the proposed algorithm in phase-shift keying (PSK) modulation and quadrature amplitude modulation (QAM) modes. The simulation results show that compared with existing advanced algorithms, the proposed algorithm can achieve performance advantages of about 1 dB and 2 dB in QPSK modulation mode and 16QAM modulation mode, respectively. [ABSTRACT FROM AUTHOR]

Brain-computer interface (BCI) provides a platform for humans to communicate using Electroencephalogram (EEG) signals by converting them into commands that can be used by the output device to perform the desired tasks. This paper focuses on the identification of vowels from EEG signals. First, a dataset of EEG signals has been created for the identification of vowels by collecting data using a 14-channel EEG device Emotiv- epoc+ from 16 subjects. Then, a deep learning-based model is proposed using a multi-headed Convolutional Neural Network for feature extraction and classification of imagined speech of vowels. Butterworth lowpass and bandpass filter of order five are implemented for denoising and sub-banding of the EEG signals which are further pre-processed using Hilbert Huang Transform. The model has achieved an average accuracy of 97.67% with a five-fold cross-validation technique using all six sub-bands of the EEG signals. The model has achieved an average precision and recall of 95.54% and 95.11% respectively. The proposed model is statistically tested using the Mann-Whitney U test and paired t-test with a p-value less than 0.05. [ABSTRACT FROM AUTHOR]

Correlation filter-based video object-tracking algorithms have gained widespread attention due to their efficiency and excellent tracking performance. However, traditional correlation filtering tracking algorithms possess several limitations. (1) They extract image features only by using rule sampling, which ignores the shape information of the target, resulting in insufficient discriminative power of the features. (2) They focus only on the difference between the background and the object, ignoring the effects of more challenging intra-class interference. (3) They do not further evaluate the reliability of the optimal candidate samples, resulting in easy failure in occlusion scenarios. This paper proposes an asymmetric background-aware correlation filter method for object tracking with deep features to solve the above limitations. First, an asymmetrical background-aware sampling method based on the shape information of the object is proposed. This sampling method significantly differs from the symmetrical sampling method in the traditional correlation filter framework. By exploring the shape information of the object, improving the otherness between the background and object samples is easy, thus suppressing the intra-class interferences. Second, deep neural networks are introduced in the correlation filter framework to extract the deep object features and a spatio-temporal regularization factor is adaptively assigned to suppress the boundary effects, intra-class distractors and aberrance between frames. Finally, a multi-modal object pool is constructed to evaluate the optimal candidate sample in each frame. This template pool fully exploits object diversity and solves the tracking drift and failure caused by invalid appearance changes in scenes such as occlusion and vigorous motion scenarios. To validate the effectiveness of the proposed method, it was compared with the state-of-the-art methods on public datasets. The experimental results show that the tracking performance of the proposed method is competitive. [ABSTRACT FROM AUTHOR]

Persymmetric structure has been utilized in space-time adaptive processing for heterogeneous environment, which leads to some detection methods based on persymmetric structure, such as persymmetric adaptive matched filter (PS-AMF). However, when the sample support is extremely limited, these methods still suffer the serious degradation in detection performance due to the large error in estimating covariance matrix. In this paper, an asymptotically optimal convex linear combination between transformed sample covariance matrix and identity matrix is considered herein to improve PS-AMF. The asymptotically optimal convex linear combination is conducive to diminishing the estimate error in estimating the transformed covariance matrix by weighting the transformed sample covariance matrix and identity matrix. Furthermore, the asymptotically optimal convex linear combination is improved by the reutilization of the convex linear combination, and the corresponding coefficients are derived. Then, the detection performance of the proposed method is approximately analyzed. At last, numerical simulations show that the proposed method performs well, compared with its counterparts, when the sample support is extremely limited. [ABSTRACT FROM AUTHOR]

Time-frequency methods, which allow us to observe the seismic signal in time and frequency domains simultaneously, can efficiently discern the seismic response features of geological bodies in different frequency bands. In this paper, a newly developed time-frequency method named second order multi-synchrosqueezing transform (SMSST) is introduced to analyze seismic data. SMSST combines the second-order instantaneous frequency estimation and the iterative idea of multi-synchrosqueezing transform to approximate the true instantaneous frequency trajectory. It solves the problems of non-reassignment points and the sensitivity to noise at the same time. Synthetic example is employed to validate the superiority of SMSST in time-frequency analysis. Results obtained from four different methods show that SMSST offer the best performance in terms of energy-concentration and noise robustness. Applications on field data further demonstrate that SMSST-based method is suitable and effective to highlight geological information and stratigraphic features compared to the previous techniques. The capabilities of SMSST for accurate reservoir localization with high resolution, making it a promising and potential tool in seismic interpretation. [ABSTRACT FROM AUTHOR]

Real-time seismic signals are intertwined with different types of noises during the generation, acquisition, and transmission process. The enhanced data with high resolution assists to interpret and analyze records more accurately. In this paper, we propose a mathematical approach based on recently developed geometrical mode decomposition (GMD) and adaptive self-guided filter (ASGF) to attenuate noise from two-dimensional seismic data. GMD is first applied to decompose the 2D seismic data into a number of band-limited intrinsic mode functions. This algorithm is capable of separating the linear and non-linear seismic events into linear modes and optimizing the linear patterns within amplitude frequency modes. The noisy modes are selected and attenuated by an adaptive self-guided filter. The GMD method is experimentally verified with a strong theoretical background to address the directional features of the image. ASGF is an exceptional edge-preserving filter and hence the hybrid algorithm could utilize the advantages of both methods. Higher the signal-to-noise ratio, improving the resolution of the image and preserving the directional properties and edges of the seismic events are the paramount characteristics of the proposed model. The simulating results on both synthetic and real seismic data proved the technique is more promising compared to the existing methods. [ABSTRACT FROM AUTHOR]

Generating Adversarial Network based on style transfer is an effective method to expand sample data. Nonetheless, an urgent issue that demands resolution is the fusion of cross-material defects and backgrounds to generate high-quality defect samples. In this paper, we propose the High-quality Matching Transfer Generative Adversarial Network (HMTGAN), an innovative framework. This network is based on the principles of CycleGAN and StyleGAN, specifically designed for advanced image generation and migration tasks. The network model incorporates multidimensional latent variables within the input image. It employs a unique method of latent space random sampling to effectively combine source defects and background materials. Additionally, a novel loss function is designed, leveraging the straight-through gradient with discrete random variables back-propagation, to effectively compare the binary defects between the source and synthetic images. Following this process, some realistic-looking defect samples are generated. We obtained a KID score of 120.76 and a FID score of 0.16, along with a Class IOU score of 0.11 on the crack defect dataset, by guiding the image generation through training style transformations. We conducted defect migration experiments on various textured surfaces to explore crack generation and migration. Extensive experiments on aluminum plate defect and crack datasets show that our method achieves state-of-the-art performance for image generation. [ABSTRACT FROM AUTHOR]

The system identification under non-negative constraints problem is a common and important one in the real-life problems. A non-negative least mean squares algorithm was proposed to address such problems. However, it suffers from slow and unbalanced convergence. Motivated by the introduction of non-Newtonian gradient in least mean squares algorithm can accelerate the convergence. In this paper, an efficient non-negative least mean squares algorithm based on q-gradient is developed. At first, the q-gradient of the mean square error cost function is derived from the definition of Jackson's derivative. Then, the modified algorithm is developed by replacing the conventional gradient with the q-gradient on a fixed-point iteration scheme using Karush-Kuhn-Tucker condition. Simulations for stationary and non-stationary system identification are conducted to illustrate the effectiveness of the developed algorithm for constrain the system parameters to be non-negative and accelerate the convergence speed. Besides, it can balance the convergence rate for widely distributed parameters, especially for the convergence of small parameters. For practical application, the accuracy of system identification when the order of the adaptive filter is larger than the order of the system is also verified by the simulation. • An efficient non-negative least mean squares algorithm based on q-gradient is developed. • The algorithm builds on a fixed-point scheme with KKT conditions using q-gradient of cost function. • The algorithm can improve the convergence speed and the balance of the convergence of parameters. [ABSTRACT FROM AUTHOR]

This paper presents a new algorithm for reconstructing time-varying graph signals using spatiotemporal feature learning. We introduce a time series analysis method to capture the temporal stationarity of graph signals and propose a reconstruction model based on spatiotemporal coupling features. However, prior knowledge of the temporal stationarity of graph signals is required for this task. To address the issue, we analyze the spatio-temporal coupling feature in a graph-spectral view and design a learning framework to capture only the temporal feature based on the sampled signals. We reconstruct the graph signals using the proposed reconstruction model and feature-learning framework by solving an unconstrained optimization problem consisting of data fidelity and two regularization terms. Numerical results using both synthetic and real-world datasets demonstrate the superiority of the proposed reconstruction algorithm over existing methods. [ABSTRACT FROM AUTHOR]

To solve direction-finding problems for nested arrays under conditions of impulse noise and gain–phase errors, this paper proposes a direction-of-arrival (DOA) estimation method. First, a correntropy with an adaptive kernel length combined with an infinite norm of a variable weight value is proposed to suppress impulse noise. Then, the auxiliary array elements are used to realize the initial calibration of the amplitude and phase errors. On this basis, a DOA estimation method without a spectral peak search is proposed; this method is robust for arrays that still have residual gain–phase errors after calibration. Our method overcomes the difficulties of DOA estimation for nested arrays under conditions of gain–phase errors and impulse noise. Compared with traditional methods, the proposed method does not require prior knowledge, does not require a spectral peak search and has high superiority and robustness; a series of experiments have verified its superiority. In addition, a Cramér‒Rao bound (CRB) for angle estimation under conditions of gain–phase errors and impulse noise is proposed in this paper. [ABSTRACT FROM AUTHOR]

The paper presents a predictive filter based frequency estimation algorithm (PF-FE) for multiple complex sinusoids corrupted by additive white Gaussian noise. The proposed frequency estimator combines the noise suppression capability of an FIR filter with a least squares approach minimizing the sum of the squared errors of a prediction filter. The FIR filter coefficients are chosen such that the sinusoidal components are predicted with minimum distortion. Computer simulation results are given to contrast the performance of the proposed PF-FE method with six other state-of-the-art frequency-estimators and the Cramer-Rao lower bound. Furthermore, the paper provides a complexity analysis of the PF-FE method and compares it with those of the conventional MUltiple SIgnal Classification algorithm and some selected state-of-the-art methods from the literature. Performance of the proposed PF-FE method is validated with effective mean square error (EMSE) plots for model orders of three and five and under different observation lengths and signal-to-noise ratio values. Finally, the paper demonstrates how EMSE varies for the proposed method when the prediction filter length is increased from 20 to 75. After extensive simulations, it is observed that when filter length is larger than half the observation length the EMSE will start to degrade. [ABSTRACT FROM AUTHOR]

This paper investigates Filter Bank MultiCarrier (FBMC) modulation implemented with frequency spreading and polyphase network-based in terms of the introduced quantization noise. As FBMC is considered one of the future candidates for 5G/6G communication systems due to its advantageous spectral properties, the introduced quantization noise in the implementation is an essential design criterion. Analytical expressions for fixed- and floating-point Quantization Noise Power (QNP) in FBMC transmitter schemes are given. Based on the results, it can be stated that the total QNP depends on the number of carriers, overlapping symbols, and the resolution of the quantizer. The results are verified through simulations. Estimating the quantization noise in FBMC systems in the function of the selected bit resolution and keeping it at an acceptable level is an essential design step. The results can be directly employed in the preliminary hardware design of FBMC transmitters, where the choice of the arithmetical units and the bit resolution is a key factor. • Different FBMC transmitter architectures are compared in terms of quantization noise. • Analytical formulas for the QNP at the output of FBMC transmitters are derived. • The effect of the Quantization noise on the PSD of the FBMC signal is analyzed. [ABSTRACT FROM AUTHOR]

Voice activity detection (VAD) is essential for multiple microphone arrays processing, in which massive potential devices, such as microphone devices for far-field voice-based interaction in smart home environments, will be activated when sound sources appear. Therefore, the VAD can save a lot of computing resources in massive microphone arrays processing for the sparsity in sound source activity. However, it may not be feasible to obtain an accurate VAD in harsh environments, such as far-field, time-varying noise field. In this paper, the long-term speech information (LTSI) and the log-energy are modeled for deriving a more accurate VAD. First, the LTSI can be obtained by measuring the differential entropy of long-term smoothed noisy signal spectrum. Then, the LTSI is used to get labeled data for the initialization of a Gaussian mixture model (GMM), which is used to fit the log-energy distribution of noise and (noisy) speech. Finally, combining the LTSI and the GMM parameters of noise and speech distribution, this paper derives an adaptive threshold, which represents a reasonable boundary between noise and speech. Experimental results show that our VAD method has a remarkable improvement for a massive microphone network. [ABSTRACT FROM AUTHOR]

This paper presents a novel closed-form analytical expression for Riesz fractional order derivative in the Fractional Fourier domain. The expression is obtained in the terms of higher transcendental functions such as Parabolic Cylinder Function as well as Confluent Hypergeometric Function. The presented work is analyzed in the discrete domain by using the properties of Discrete Fractional Fourier Transform (DFrFT). The proposed algorithm is capable of preserving the texture and edge information without any phase distortion. The design example discussed in the paper shows the efficacy of the proposed algorithm for a signal with high frequency chirp noise. The design flexibility of the proposed approach is confirmed due to the fact that it provides an optimal value of performance metrics such as Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) corresponding to the variation of the fractional order of Riesz derivative and fractional parameter in the rotation angle of Fractional Fourier Transform (FrFT). The proposed algorithm provides better results in terms of minimum RMSE of 0.115136 and MAE of 0.094223 for the optimal fractional order of 0.43 at a rotation angle of 0.45 π. [ABSTRACT FROM AUTHOR]

Sensor position and velocity uncertainties are known to be able to degrade the source localization accuracy significantly. This paper focuses on the problem of locating multiple disjoint sources using time differences of arrival (TDOAs) and frequency differences of arrival (FDOAs) in the presence of sensor position and velocity errors. First, the explicit Cramér–Rao bound (CRB) expression for joint estimation of source and sensor positions and velocities is derived under the Gaussian noise assumption. Subsequently, we compare the localization accuracy when multiple-source positions and velocities are determined jointly and individually based on the obtained CRB results. The performance gain resulted from multiple-target cooperative positioning is also quantified using the orthogonal projection matrix. Next, the paper proposes a new estimator that formulates the localization problem as a quadratic programming with some indefinite quadratic equality constraints. Due to the non-convex nature of the optimization problem, an iterative constrained weighted least squares (ICWLS) method is developed based on matrix QR decomposition, which can be achieved through some simple and efficient numerical algorithms. The newly proposed iterative method uses a set of linear equality constraints instead of the quadratic constraints to produce a closed-form solution in each iteration. Theoretical analysis demonstrates that the proposed method, if converges, can provide the optimal solution of the formulated non-convex minimization problem. Moreover, its estimation mean-square-error (MSE) is able to reach the corresponding CRB under moderate noise level. Simulations are included to corroborate and support the theoretical development in this paper. [ABSTRACT FROM AUTHOR]

The problem of detecting a subspace signal embedded in subspace Gaussian interference and thermal noise is studied in this paper. In this problem, both the signal-independent and signal-dependent interferences are assumed to be present, therefore the overall interference subspace covers the signal subspace. The approach of this paper extends previous works involving either of those two kinds of interferences. A set of secondary data containing only interference plus noise is employed to estimate the interference covariance matrix and the noise power. Three new detectors are designed via the generalized likelihood ratio (GLR), Rao and Wald tests, respectively. Their probabilities of false alarms (PFAs) and detections are analytically derived. The PFAs show that the new detectors have the constant false alarm rate (CFAR) property against the interference and noise. Numerical results show that the new detectors outperform their counterparts for the studied problem. Furthermore, the new detectors are less sensitive to the secondary data size and to the mismatched subspace signal than some other detectors, such as the GLR detector (GLRD), the adaptive matched filter (AMF), the adaptive subspace detector (ASD), etc. [ABSTRACT FROM AUTHOR]

• A new method is proposed to dynamic texture analysis. • The video is modeled into directed networks. • The dynamic texture is characterized by the diffusion over the networks. • The experiments on three dynamic texture datasets validate its effectiveness. Dynamic texture is a field of research that has gained considerable interest from computer vision community due to the explosive growth of multimedia databases. In addition, dynamic texture is present in a wide range of videos, which makes it very important in expert systems based on videos such as medical systems, traffic monitoring systems, forest fire detection system, among others. In this paper, a new method for dynamic texture characterization based on diffusion in directed networks is proposed. The dynamic texture is modeled as a directed network. The method consists in the analysis of the dynamic of this network after a series of graph cut transformations based on the edge weights. For each network transformation, the activity for each vertex is estimated. The activity is the relative frequency that one vertex is visited by random walks in balance. Then, texture descriptor is constructed by concatenating the activity histograms. The main contributions of this paper are the use of directed network modeling and diffusion in network to dynamic texture characterization. These tend to provide better performance in dynamic textures classification. Experiments with rotation and interference of the motion pattern were conducted in order to demonstrate the robustness of the method. The proposed approach is compared to other dynamic texture methods on two very well known dynamic texture database and on traffic condition classification, and outperform in most of the cases. [ABSTRACT FROM AUTHOR]

Recent years have witnessed the explosive growth of multimedia applications over networks and increasingly high requirements of consumers for multimedia signals in terms of quality of experience (QoE). Effective and efficient yet energy-saving saliency detection model and quality prediction method are eagerly desired, since they play critical roles in raising users' QoE and promoting the progress of green multimedia communication. Current studies of saliency detection and quality evaluation are far from ideal yet. In this paper we investigate the influence of complexity on visual saliency and quality. Complexity is an essential concept in human perception to visual stimulus, but it is substantially abstract and hard to be endowed with a clear definition. We suppose that brain systematically combines global and local features during the whole process of human perception. Global features lead a dominant position in seeking salient areas under the condition that image complexity is high, namely without obviously isolated foreground objects, whereas local features play a key role in an opposite situation. With this consideration, this paper establishes a novel framework for detecting visual saliency based on image complexity estimation before complexity-adaptive merging of global and local features. Furthermore, the concept of complexity is deployed for blind photographic image quality assessment (IQA) by means of saliency-based weighting. Features which refer to contrast, artifacts, brightness and natural scene statistics (NSS) are modified and integrated to derive a blind IQA model and predict the quality of photos. Based on the above two technologies, this paper introduces smart phones as mobile terminals, cloud platforms for speed-up and energy-saving, and wireless networks for transmission, and provides a practical mobile multimedia application. Comparative experiments validate that, within this application system, our proposed saliency detection model and blind photographic IQA method implement better than existing relevant competitors in terms of effectiveness and efficiency comparison. [ABSTRACT FROM AUTHOR]

This paper studies the causes and distribution of reconstruction errors in subtractive dither structure and provides solution to minimize these errors. We present the generation of reconstruction errors in two types of quantization, namely mid-riser and mid-tread, with their corresponding distributions. We then construct a 12-bit pipeline analog to digital converters (ADC) simulation model and use a 16-bit digital to analog converter (DAC) to convert the dither sequence into analog signals. We perform statistical analysis on the resulting data, verify the accuracy of the distribution, and find that reconstruction error only occurs when the dither signal accuracy surpasses that of the ADC. Additionally, we find that mid-tread quantization has the lowest probability of reconstruction error. Finally, we analyze the impact of reconstruction error on the signal to noise ratio (SNR) and spurious free dynamic range (SFDR) at various input frequencies. Our findings show that the reconstruction errors' effect on the SFDR and SNR of the input signal is less than 1 dB. [ABSTRACT FROM AUTHOR]

The interrupted-sampling repeater jamming (ISRJ) is a mainstream jamming method in electronic warfare, simple yet implemented. With the rapid development of anti-ISRJ, due to the defects of ISRJ and the difficulty in ensuring real-time response, there is an urgent need to propose a new ISRJ-based jamming method based on simultaneous transmission and reception (STAR) technology. An interrupted-sampling multi-strategy forwarding jamming (ISMSFJ) with amplitude constraints is proposed by utilizing the characteristics of interrupted sampling in this paper. By modulating different interrupted sampling pulses on the intercepted signal and accurately controlling the delay time and weighting coefficient, it not only destroys the space distribution law of false targets, shortens the response time of the jammer, but also enhances the jamming amplitude or broadens the jamming range. For radar receivers, the jamming overlaps with the time-frequency information of the radar echo signal, and the time-frequency analysis anti-jamming algorithm fails. It can effectively resist the Fractional Fourier Transform anti-jamming algorithm for interpulse and intrapulse frequency-agile signals. The validity of the jamming technology is verified by theoretical analysis and simulation. [ABSTRACT FROM AUTHOR]

Recently, deep learning methods have achieved excellent performance in single image super-resolution (SISR), but most existing methods suffer from heavy computational costs and memory storage. To address this problem, some lightweight SR methods have been proposed, among which convolutional neural network (CNN) with attention mechanisms has received increasing attention. However, the existing CNN-based lightweight network and attention mechanism still have redundancy in feature extraction and enhancement. This paper presents Efficient Self-calibrated and Hierarchical Refinement Network (ESHR) for Lightweight Super-Resolution that includes three effective designs. First, we propose a more effective feature extraction basic module called Self-calibrated Residual Block (SRB), which can achieve superior performance when replacing other common convolution modules such as traditional convolution, BSConv and DSConv. Second, we propose a more lightweight spatial attention mechanism called Squeeze-Excitation Spatial Attention (SESA), which can interact and enhance features more effectively in spatial dimensions to improve the representation ability of the model. Third, we design an Efficient Hierarchical Refinement Block (EHRB) to integrate and utilize multi-level features, which can achieve better performance than single-level refinement without increasing the number of parameters. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art CNN-based methods quantitatively and qualitatively. [ABSTRACT FROM AUTHOR]

A delay dynamical system can fold a feedforward neural network into one nonlinear neuron and multiple delay loops under the non-Von Neumann structure, greatly decreasing the hardware requirements. In this paper, we transform the folded-in-time DNN (Fit-DNN) into a folded-in-time RNN (Fit-RNN) and derive the backpropagation algorithm for it. The performance of the folded reservoir computing (DRC), Fit-DNN, and Fit-RNN is compared on time-series prediction and classification tasks, respectively. The impact of virtual node separation on the performance of Fit-RNN is analyzed. The limits of Fit-RNN's capabilities in conducting effective time-series predictions were determined based on the NARMA-X series tasks. We calculate the 5-order information processing capacity (IPC) of DRC and Fit-RNN on NARMA10. The results indicate that Fit-RNN has almost the necessary information transformation ability for the task. In the recognition tasks conducted separately on spoken digits and speakers, we investigate the impact of the number and separation of virtual nodes on the recognition capability of the three folded networks. The results demonstrate that Fit-RNN shows more promise in handling long sequences and large-scale recognition tasks. The recognition accuracy is further enhanced by increasing the number of virtual nodes and node separation. Furthermore, the introduction of two common types of noise in the speaker recognition task environment further highlights the potential of Fit-RNN in practical applications. Furthermore, two common types of noise are introduced into the speaker recognition task, further highlighting the potential of Fit-RNN in practical applications. [ABSTRACT FROM AUTHOR]

The security and protection of digital content transmitted over a network are critical concerns due to the ease with which various types of data can be transformed into text, images, audio, and video forms. This vulnerability makes them susceptible to unauthorized modifications by hackers. To address this issue, several techniques have been developed, and one such approach is image encryption and watermarking. Watermarking offers solutions for data authentication, copyright protection, and document ownership problems. This paper introduces a new algorithm based on Sparse Approximation (SA), Quantum Encryption (QE), and a measurement matrix to achieve watermark encryption, as well as its secure transmission by embedding it in a host image. In this, the watermark image is divided into four sub-images, and then sparsification is performed in the transform domain. To further enhance security, the sparse data is transformed in the Discrete Wavelet Transformation (DWT) domain, and a sparse coefficient exchange procedure is applied. Next, we generate a measurement matrix, and the row and column vector of this matrix are filled by sequences which are generated by a quantum logistic map, and then perform encryption as well as data scrambled process. During encryption, updated pixel location is calculated based on a sparse sub-band coefficient exchange procedure. Based on this procedure, we exchange the sparse coefficient between the sub-images. Finally, we generate an encrypted watermark through a series of inverse sparsification, DWT, and sampling. These steps collectively fortify the algorithms against various attacks, making them robust. The encrypted watermark information is used for authentication purposes and is embedded into the host image using DWT and Singular Value Decomposition. The results are presented in various tabular and graphical forms, showcasing the effectiveness of the proposed method. The experimental results show that the proposed techniques resist several attacks such as enhancement attacks, noise impact attacks, and geometrical attacks. A comparative analysis is also provided, comparing the results with existing approaches. [ABSTRACT FROM AUTHOR]

This paper presents an advanced region growing method for precise left atrium (LA) segmentation and estimation of atrial wall thickness in CT/MRI scans. The method leverages a Gaussian mixture model (GMM) and fractal dimension (FD) analysis in a three-step procedure to enhance segmentation accuracy. The first step employs GMM for seed initialization based on the probability distribution of image intensities. The second step utilizes fractal-based texture analysis to capture image self-similarity and texture complexity. An enhanced approach for generating 3D fractal maps is proposed, providing valuable texture information for region growing. In the last step, fractal-guided 3D region growing is applied for segmentation. This process expands seed points iteratively by adding neighboring voxels meeting specific similarity criteria. GMM estimations and fractal maps are used to restrict the region growing process, reducing the search space for global segmentation and enhancing computational efficiency. Experiments on a dataset of 10 CT scans with 3,947 images resulted in a Dice score of 0.85, demonstrating superiority over traditional techniques. In a dataset of 30 MRI scans with 3,600 images, the proposed method achieved a competitive Dice score of 0.89±0.02, comparable to Deep Learning-based models. These results highlight the effectiveness of our approach in accurately delineating the LA region across diverse imaging modalities. [ABSTRACT FROM AUTHOR]

Recently, some graph-based methods have been proposed for multi-model fitting. These methods usually construct full-connected graphs between any pairs of data for model fitting. However, the constructed graphs tend to contain many invalid edges, especially for input data corrupted by a large number of noise and outliers, which may lead to suboptimal fitting results. In this paper, we propose a novel multi-model fitting method based on neighborhood graph structure consistency (NGSC), which preserves the local neighborhood structures of potential inliers to construct an effective graph for robust model fitting. Specifically, by exploiting the motion and preference information of input data, we adaptively generate the local-neighbors for each vertex of the graph. To enhance the effectiveness of graph construction, we propose an effective guided sampling strategy to improve the accuracy of preference correlation measurement. In addition, we propose a novel graph clustering algorithm to effectively detect the connected components of the graph (i.e., connected subgraphs) corresponding to model instances, for model selection. Overall, the key elements (i.e., graph construction, graph partition, and subgraph detection) are original. Extensive experimental results on several challenging datasets show that the proposed NGSC achieves more accurate fitting results than other several state-of-the-art fitting methods. [ABSTRACT FROM AUTHOR]

This paper introduces a novel localized Gini index (GI) designed for the assessment and optimization of time-frequency distributions (TFDs). This approach employs the localized Rényi entropy (LRE) to quantify the local number of signal components, providing a unique perspective on interpreting time and frequency slice segments while considering the presence of auto-terms or cross-terms. The results demonstrate the effectiveness of the proposed LRE-based GI in overcoming the limitations of conventional concentration and sparsity measures, particularly in scenarios with varying component numbers and missing essential components that may arise during TFD reconstruction using the compressive sensing method. Furthermore, it proves as an efficient objective function in meta-heuristic optimization, ensuring the preservation of auto-terms and a significant enhancement in overall TFD quality. The performance of the proposed LRE-based GI is demonstrated on noisy synthetic and real-life signals. • Current measures possess limitations when assessing sparse TFDs. • A novel Gini index for assessing sparse TFDs with variable components is introduced. • The localized Rényi entropy is utilized in the novel Gini index. • Meta-heuristic optimizes by maximizing LRE-based Gini index. [ABSTRACT FROM AUTHOR]

This paper introduces an innovative methodology to establish consensus among distributed information processing nodes (DIPNs) in the context of multi-target tracking (MTT) within environments characterized by resource constraints and communication uncertainties. Through the integration of the event-triggered (ET) strategy with a consensus-based algorithm, existing approaches foster consensus among DIPNs while simultaneously conserving communication resources. Nonetheless, a systematic investigation into the comprehensive analysis of data reliability from each node has not been conducted. Combining anomalous data resulting from communication uncertainties with other data on an equal footing leads to inaccurate results. To address this issue, we apply the multiple-model algorithm, assigning consensus weight to each DIPN based on the motion model distribution of the same target observed by different DIPN. Additionally, we introduce an auxiliary ET marker, considering the divergence in the motion model distribution between two consecutive moments of a certain target. This marker assists in determining whether local information must be transmitted to other DIPNs. The proposed approach yields more accurate and congruent output results from each DIPN in comparison to conventional methods, given the same triggering frequency. Numerical simulations demonstrate the efficacy of the suggested approaches in a distributed MTT scenario. • We apply the multiple-model generalized labeled multi-Bernoulli filter. • We combine event-triggered strategy with consensus approach to reduce communication. • We check data reliability from each node to improve multi-target tracking accuracy. • We suggest an additional event-triggered mark to locate the message for transmitting. [ABSTRACT FROM AUTHOR]

The presence of noise is a pervasive issue that significantly impacts the performance of adaptive filtering algorithms. To address this challenge, the bias compensation technique has recently emerged, involving the incorporation of an additional term into the update equation. Despite the apparent simplicity of the bias-compensated least-mean-squares algorithm, conducting a comprehensive theoretical analysis of its performance is a complex task. In this paper, we undertake an exact expectation analysis to demonstrate the asymptotic unbiasedness of the algorithm, even when disregarding the commonly assumed theory of independence between adaptive coefficients and input data. Furthermore, to improve the understanding of the algorithm's stability, we employ a stochastic model that assumes independence between the radial and angular distributions of the input vector. The resulting model is intricate, necessitating heuristic approximations to derive practical insights. Notably, our analysis reveals that the upper bound on the step size of the bias-compensated least-mean-squares algorithm consistently remains smaller than that of the least-mean-squares algorithm. These findings gain robust support from extensive simulations. [ABSTRACT FROM AUTHOR]

In this paper, we design a low-complexity direction-of-arrival (DOA) estimation algorithm based on the unitary approximate message passing (UAMP) and Bernoulli-Gaussian (BG) prior. We first show that the estimation of DOA can be transferred into a sparse signal recovery problem, where we turn to UAMP with damping technique to solve this problem. Furthermore, we assume the BG prior on the sparse vector to be estimated, resulting in the fast estimation of the positions of non-zero elements. Moreover, expectation maximum (EM) is leveraged to automatically learn the BG parameters. Compared to the state-of-the-art UAMP-based algorithm with sparse Bayesian learning (SBL), the proposed approach can achieve the same DOA mean square error (MSE) performance with a much faster convergence speed. [ABSTRACT FROM AUTHOR]

To improve the robust characteristic and filtering accuracy of the traditional adaptive filtering algorithms (AFAs) under censored regression (CR) model in impulsive interference environments, some robust CR-type AFAs (R-CR-AFAs) have been developed. However, these exiting R-CR-AFAs may suffer from high steady-state misalignment in parameter estimation scenario. Therefore, in this paper, we first construct a unified framework to develop a class of R-CR-AFAs by adopting various robust strategies with error nonlinearities as cost functions. Then the mean and mean-square stability ranges and steady-state convergence behavior of the R-CR-AFAs are analyzed via resorting to some frequently-utilized assumptions and lemmas. In particular, a novel censored regression generalized modified Blake-Zisserman adaptive filtering (CR-GMBZ-AF) algorithm is presented to achieve higher steady-state estimation accuracy contrasted to related R-CR-AFAs benefiting from the advantages of the newly proposed GMBZ criterion based on the unified framework. Numerical simulation experiments manifest that the developed CR-GMBZ-AF algorithm arrives at better learning performance in comparison with state-of-the-art algorithms and the theoretical steady-state model demonstrates good agreement with simulation results under parameter estimation and acoustic echo cancellation scenarios. [ABSTRACT FROM AUTHOR]

Pedestrian target tracking is a key area in computer vision. This paper tackles the problem of low accuracy in conventional pedestrian target tracking by introducing an Adaptive Mixture Gaussian Particle Filter (AMGPF), which integrates Variational Mode Decomposition and autoregressive models. This method builds upon the traditional particle filter algorithm and introduces several improvements. First, we enhance the particle filter by introducing a mixture Gaussian model to obtain the posterior probability distribution of particles. This representation allows each particle to capture a mixture of possible states, alleviating the problem of particle degeneracy commonly encountered in traditional particle filters. Next, we refine the iterative process of the mixture Gaussian model by employing Bayesian criteria and likelihood function convergence to adaptively control the number of iterations and the Gaussian component count. This optimization helps reduce computational load while improving the model's goodness of fit. Lastly, we introduce variational mode decomposition to process the residual signals. By decomposing and reconstructing these signals, we extract multi-modal motion features. Based on this, we employ an autoregressive model for prediction, combining the predicted trajectory with the original one to effectively enhance tracking accuracy. Experimental results demonstrate that the proposed method significantly improves the precision of pedestrian trajectory prediction and exhibits strong robustness, outperforming traditional methods. • A hybrid Gaussian model is constructed to obtain the posterior probability particle distribution, which reduces the problem of particle degradation. • The Bayesian criterion is used to control the convergence of the mixed Gaussian iterations and the likelihood function of Gaussian components, which reduces the computational complexity and improves the efficiency of the algorithm. • The residual error of the predicted trajectory is smoothed and predicted, and the prediction accuracy is significantly improved. [ABSTRACT FROM AUTHOR]

In recent years, graph convolutional networks (GCNs) have made remarkable achievements in the hyperspectral image (HSI) classification task. However, existing GCN-based methods cannot adequately encode similarity edge relationship between superpixels, and few of them use hierarchical mechanism to extract complementary features. This paper addresses these issues and proposes a hierarchical structural graph neural network with local relation enhancement (HSLRE) for HSI classification. Specifically, the features of the pixel-level graph structures are extracted and then embedded into the superpixel-level graph structure to ensure that it does not lose the fine texture features of the original HSI. Secondly, a novel hierarchical framework, which consists of multiple coarsening and refining stages, is proposed to extract multi-level features. In the first coarsening stage, the relational graph convolution (RGC) is introduced to enhance local relations and obtain discriminative features from the superpixel-level graph. In the subsequent coarsening stages, graph convolution (GC) is used to extract features. The refining stages correspond to the coarsening stages, which are used to restore the graphs to their original structures. Finally, to enhance the fluidity of feature information, the fully connected layers and two different types of graph convolutional layers are utilized to extract the linear and nonlinear features of the nodes in parallel, which are fused in a weighted way to form effective features. Experimental results on several benchmark HSI datasets illustrate the effectiveness of the HSLRE. [ABSTRACT FROM AUTHOR]

Nonlinear adaptive filters, such as Volterra and functional link adaptive filter (FLAF), face increased computational complexity with rising input dimensionality or expansion order. To enhance computational efficiency, this paper introduces a nonlinear adaptive algorithm based on Rectified Linear Unit (ReLU). Applying ReLU to a tonal signal generates components corresponding to the original frequency and even harmonics. Additionally, applying ReLU to the sum of two tonal signals yields a spectrum encompassing intermodulation distortion frequencies. These traits enable the algorithm to mitigate harmonic and intermodulation distortions while maintaining judicious computational complexity. The numerical simulation results for the identification of nonlinear systems validate the algorithm's ability to fit harmonic and intermodulation nonlinear distortions. In simulations of nonlinear echo cancellation with diverse test signals and various distortion models, our proposed algorithm consistently demonstrates competitive performance, notably excelling in music and speech scenarios. [ABSTRACT FROM AUTHOR]

Feature extraction has a significant impact on the accuracy of Content-Based Image Retrieval (CBIR) algorithms since the content of images is encoded in feature vectors. In this paper, an effective method for texture feature extraction is proposed based on local patterns. In the proposed method, first the image is formed in different scales, then the texture features are extracted from the scales of the image. Finally, extracted features of different scales are concatenated to construct the final feature vector. To evaluate the proposed method, five datasets including Corel-1k, Brodatz, VisTex, Corel-10k, STex, Caltech256, and Oliva are used. In the evaluation process, the effect of different scales and different coding schemes on retrieval precision is investigated. The proposed method is compared with existing CBIR models based on local patterns. The proposed method achieves the best precision of 64.16%, 81.66%, 88.59%, 30.63%, and 69.63%, 13.59%, and 64.10% on the Corel-1k, Brodatz, VisTex, Corel-10k, STex, Caltech256, and Oliva datasets, respectively. [ABSTRACT FROM AUTHOR]

The high altitude images captured by drones often contain small-sized objects, which pose several challenges including significant scale variations and dense clustering. To tackle these challenges in small object detection, this paper presents EM-YOLO (Efficient Switching Network and Multi-Scale Feature Fusion Reinforced YOLO for Small Object Detection), an enhanced algorithm based on the YOLOv5s architecture. Firstly, to enhance the network's adaptability to varying object feature scales, the EM-YOLO redesigns the feature extraction backbone network by using different dilation rates in convolutions within the backbone. This flexible structure allows for improved extraction of object features, enhancing the network's feature extraction capability. Secondly, additional detection layer for small objects and combines it with the dynamic detection head. This combination addresses the issue of small object information loss due to multiple down-sampling operations on the feature maps. Finally, Focal-EIoU Loss is employed to enhance regression accuracy in EM-YOLO. Extensive experiments were conducted on the VisDrone2019-DET dataset. The results of these experiments demonstrate that the improved algorithm achieved a significant increase in performance. Specifically, the algorithm achieves 43.5% and 25.1% on the mAP 0.5 and mAP 0.5 : 0.95 metrics, which are 13.9% and 8.9% higher than YOLOv5s algorithm, respectively. Furthermore, comparative experiments with other mainstream algorithms indicate that EM-YOLO outperforms them in the task of small object detection. • Solve the small object detection of complex background. • It has stronger characteristic scale adaptability. • Ease the problem of loss of small object information. • Improves the regression accuracy of small object predictions. [ABSTRACT FROM AUTHOR]

Attitude and size estimation of spacecraft components is one of the main contents of space target awareness. In this paper, a radar-camera joint observation scheme applicable to multiple spaceborne platforms is developed to measure the orientation and length of the spacecraft solar panels, which benefits from the observation diversity jointly provided by radars and cameras. Firstly, an automatic keypoint detection network is trained to extract the endpoints of the solar panels in the radar and optical images with high accuracy. Then, the projected lengths of the solar panels calculated from the extracted endpoints along with the observation equations of the radars and cameras are jointly used to obtain the orientation and actual length of the solar panels by an optimization approach. Simulated experiments are conducted to demonstrate the performance of the proposed measuring methods, which achieves length estimation errors within 0.2 m and orientation estimation errors within 1°. The proposed method can be extended to the measurement of main body and other spacecraft components having regular features. [ABSTRACT FROM AUTHOR]

Contemporary Internet-of-Things (IoT) devices in general and Internet-of-Vehicles (IoV) networks in specific are taking the benefits from using full-duplex (FD) communication with the help of relay nodes to facilitate data transmission. However, self-interference between transmit and receive antennas at the relay node may cause big trouble if self-interference cancellation (SIC) is not applied. In this paper, a rigorous analysis on the performance of FD two-way relay networks (FDTWRNs) for IoV applications under imperfect conditions is provided. Different from earlier studies on the performance of FDTWRNs where one or two imperfect factors were considered, this work simultaneously takes into account many practical issues in the implementation of the IoV networks at the physical layer, including errors in channel state information at all channel links, imperfect SIC and transceiver hardware impairment at all nodes. For relaying protocol, we consider amplify-and-forward with both fixed-gain (FG) and variable-gain (VG) protocols. The exact closed-form expressions (in terms of the system parameters) of key performance factors such as signal-to-interference-plus-noise-distortion-and-error ratio (SINER), outage probability (OP), achievable data rate (ADR), and average symbol error probabilities (ASEP) of the considered system over Rayleigh fading channels and under the combined effect of imperfect CSI, SIC, and transceiver hardware are derived mathematically using probabilistic model analysis. The correctness of the analysis is then verified by Monte Carlo simulations. Drawing upon both analytical and numerical findings, the efficacy of the FDTWRNs under consideration is meticulously and comprehensively assessed in comparison to its performance in ideal conditions characterized by one or multiple critical factors. The disparity between perfect and imperfect scenarios serves as a valuable metric for appraising the practical viability of the considered FDTWRNs when deployed in practical IoV networks. [ABSTRACT FROM AUTHOR]

Anomaly detection is distinguishing unusual objects from normal patterns. It is a complex task due to unpredictable nature of anomalies, which can appear in many forms or they can be hidden by mimicking normal behaviors in a graph structure. Such diversity makes this Deep learning approaches can solve these problems by extracting complex patterns from networks. However, addressing different forms of anomalous instances is essential for successfully implementing these approaches, as different anomaly types require further analysis. Additionally, it is challenging to interpret anomalies beforehand without focusing on every aspect of anomalies. Our objective is to propose an architecture capable of handling all types of anomalous entities by tackling challenges across various domains. In this paper, we introduce ARNAD, a novel framework that integrates three deep models to identify anomalies in graphs: graph neural network, autoencoder, and adversarial autoencoder. ARNAD approaches graph anomaly detection by utilizing the features of the deep parts, and four key elements stand out: (1) the autoencoder learns the overall graph structure and identifies highly deviated ones, (2) the graph neural network exploits graph structure to detect anomalies among the communities, (3) a fixed-size randomized neighborhood that prevents overfitting while reducing complexity (4) the adversarial autoencoder improves the robustness of the framework and discriminates anomalies. To detect anomalies, four receptive components assign risk scores to objects in the attributed network. We evaluated the framework with three synthetic datasets that simulate different behaviors of anomalies and six widely used real attributed networks. Our experimental results show that ARNAD performs competitively with other state-of-the-art models in detecting anomalous entities while minimizing false positives, demonstrating ARNAD's effectiveness in detecting graph anomalies. [ABSTRACT FROM AUTHOR]

The work presented in this paper aims at enhancing the performance of end-to-end (E2E) speech recognition task for children's speech under low resource conditions. For majority of the languages, there is hardly any speech data from child speakers. Furthermore, even the available children's speech corpora are limited in terms of the number of hours of data. On the other hand, large amounts of adults' speech data are freely available for research as well as commercial purposes. As a consequence, developing an effective E2E automatic speech recognition (ASR) system for children becomes a very challenging task. One may develop an ASR system using adults' speech and then use it to transcribe children's data, but this leads to very poor recognition rates due to the stark differences in the acoustic attributes of adults' and children's speech. In order to overcome these hurdles and to develop a robust children's ASR system employing E2E architecture, we have resorted to several out-of-domain and in-domain data augmentation techniques. For out-of-domain data augmentation, we have explicitly modified adults' speech to render it acoustically similar to that of children's speech before pooling into training. On the other hand, in the case of in-domain data augmentation, we have slightly modified the pitch and duration of children's speech in order to create more data capturing greater diversity. Data augmentation approaches helps in mitigating the ill-effects resulting from the scarcity of data from child domain to a certain extent. This, in turn, reduces the error rates by a large margin. In addition to data augmentation, we have also studied the efficacy of Gamma-tone frequency cepstral coefficients (GFCC) and frequency domain linear prediction (FDLP) technique along with the most commonly used Mel-frequency cepstral coefficients (MFCC) for front-end speech parameterization. Both MFCC as well as GFCC capture and model the spectral envelope of speech. On the other hand, application of linear prediction on the frequency domain representation of speech signal helps to effectively capture the temporal envelope during front-end feature extraction. Employing FDLP features that model the temporal envelope provides important cues for the perception and understanding of stop bursts and, at times, complete phonemes. This motivated us to perform a comparative experimental study of the effectiveness of the three aforementioned front-end acoustic features. In our experimental explorations, the use of proposed data augmentation in combination of FDLP features has shown a relative improvement in character error rate by 67.6% over the baseline system. The combination of data augmentation with MFCC or GFCC features is observed to result in lower recognition performances. [ABSTRACT FROM AUTHOR]

This paper studies the waveform design problem for reconfigurable intelligent surface (RIS) aided multifunction radio frequency (MFRF) systems, which enables detection, communication, and electronic attack simultaneously. The design purpose is to maximize the relative entropy while ensuring effective communication and jamming functionalities. To improve the practicability of the MFRF system, a constant-modulus constraint is imposed on the transmit waveforms. A cyclic method is proposed to iteratively optimize the transmit waveforms and the RIS coefficients. Specifically, to tackle the resultant non-convex problems, we resort to the minorization-maximization based method to transform the objective function into a quadratic form, then leverage alternating direction method of multipliers to seek efficient solutions. Numerical results verify the effectiveness of the proposed algorithm and show that with the aid of RIS, the target detection performance can be effectively enhanced. [ABSTRACT FROM AUTHOR]

Remaining useful life prediction (RUL) under variable operating conditions has been a hot research topic in engineering. Still, current unsupervised domain adaptive (UDA) methods all use a single metric (MK-MMD or adversarial mechanism), which somewhat improves the cross-domain RUL prediction performance. However, using MK-MMD or the adversarial mechanism alone to measure the inter-domain differences has a single perspective. At the same time, MK-MMD focuses on measuring the differences in the feature distribution layer and fails to capture the differences in the nonlinear relationship between the source and target domains. In addition, learning high-quality degraded features is also a vital issue in RUL prediction. Based on the above points, this paper proposes an attention mechanism multiscale capsule network cross-domain lifetime prediction method (DI-MCNAM) based on domain invariant properties. First, the attention mechanism multiscale capsule network (MCNAM) extracts the deep degradation features of the weighted data. Second, two modules, multilevel domain adaptation and domain classifier, are integrated to reduce inter-domain differences from different perspectives. The maximum information coefficient (MIC) is introduced in the multilevel domain adaptation module. MIC can learn the nonlinear relationship between source and target domain data over time separately, combined with MK-MMD, to reduce the inter-domain feature differences from different levels. Domain classifiers discriminate the confusing domains and reduce inter-domain differences by minimizing the classification power. The performance of the proposed model is validated using publicly available datasets and compared with existing popular methods, and the final results show that the proposed method has high prediction accuracy. [ABSTRACT FROM AUTHOR]

In this paper we consider the problem of estimation of scalar field distribution (e.g. pollutant, moisture, temperature) from noisy measurements collected by unmanned autonomous vehicles such as UAVs. The field is modelled as a sum of Fourier components/modes, where the number of modes retained and estimated determines in a natural way the approximation quality. An algorithm for estimating the modes using an online optimisation approach is presented, under the assumption that the noisy measurements are quantized. The algorithm can also be used to estimate time-varying fields. Simulation studies demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

The conventional steering vector (SV) based beamformers such as the minimum variance distortionless response (MVDR) can suffer from severe performance degradation due to the signal cancellation in the presence of correlated interferences. A partially correlated interference can be decomposed into two components, the coherent component and the other uncorrelated with the desired signal. The SVs of the desired signal and the coherent interference components compose the composite steering vector (CSV). In this paper, a beamforming method is presented that can achieve the maximum signal-to-interference-plus-noise ratio (SINR) via the use of the CSV and the interference covariance matrix (ICM) formed by the uncorrelated interference components only. The estimates of the CSV and the ICM are extracted from the sample matrix through oblique projections. In the estimation process, neither the interference powers nor the desired signal power is estimated, though the direction estimation is carried out using the multiple signal classification (MUSIC). We make theoretical performance analysis, which shows that the CSV based method is more effective as the correlation between the desired signal and the interferences increases and as the desired signal power becomes small. Simulation results demonstrate that it not only converges very quickly to the maximum SINR but also is robust to pointing errors. [ABSTRACT FROM AUTHOR]

Radio frequency fingerprint (RFF) identification has emerged as a promising technique for physical-layer identity authentication. However, most RFF identification schemes tend to make the restrictive closed-set assumptions and focus solely on RFF classification. In this paper, we propose a novel open-set RFF identification scheme, with both device classification and rogue device detection capability. Moreover, to reduce the open space risks and mitigate the impact of channel noise, an RFF extraction scheme based on Fourier-based synchrosqueezing transform (FSST) and supervised contrastive learning (SCL) is proposed to acquire more concentrated and distinctive features. The performance of the proposed scheme is experimentally demonstrated using the datasets from 11 commercial wireless network interface controllers (WNICs) as well as open-source datasets. The results verify the significant improvements in high-openness and low-SNR scenarios, with high identification accuracy of 82 %, at a low SNR of 5 dB, while the openness can be as high as 21 %, and an accuracy of approximately 100 % when SNR>20 dB. [ABSTRACT FROM AUTHOR]

This paper proposes low-biased probabilistic multipliers and applies the proposed designs to the inference stage of convolutional neural networks (CNNs). Highly inaccurate compressors in the probabilistic multiplier make the error distribution unbalanced and produce significant relative errors. We describe design rules for applying different compressors to the probabilistic multiplier. Besides, the proposed design rules enhance the error characteristics by determining the input values of each compressor, suppressing the significant relative error. Notably, motivated by the design rules, we propose a novel inaccurate 4:2 compressor suitable for our new design. The design rules and proposed 4:2 compressor are adopted in 8-bit probabilistic multipliers. The error analysis shows that the new designs can enhance error characteristics without increasing hardware costs. Furthermore, the proposed approximate multipliers are applied to the convolutional layers of CNNs. We describe the fine-tuning scheme to reduce the retraining time of approximation-aware training. The experiments show that proposed designs can produce acceptable classification results compared with those using the floating-point and 8-bit exact multiplications on CIFAR and ImageNet datasets. [ABSTRACT FROM AUTHOR]

There has been a continued need for technology to prevent the illegal use of music in online transactions and protect the rights of copyright holders. Recently, the types of attacks have become very diverse due to various online environments and content production methods. Therefore, an adaptive protection method aware of content and attacks is needed. This paper proposes a robust watermarking algorithm of a blind method to protect audio copyright in online streaming services. After detecting the transient points of the audio and processing the signals, they are used as reference triggers for frequency conversion. Transient triggers should be determined to be detected even after an attack. Audio samples are converted into the fast Fourier transform (FFT) frequency coefficients, and a watermark is embedded using the quantization index modulation (QIM) method. Various user-defined parameters can be adjusted during the watermark embedding process. Such parameters may be determined through experiments or an optimization process by targeting both the quality of the watermarked audio and the bit error rate (BER) of the watermark extracted after the attack. A watermark can be extracted using a similar process using parameters used in watermark embedding. If there is no information about the parameters used in embedding, the watermark can be extracted by tracking the minimum BER of the extracted watermark. Compared to the existing methods, the proposed watermarking technique showed excellent performance against most attacks and superior performance for filtering, shifting, and moving picture expert group (MPEG)-2 audio layer 3 (MP3) compression attacks. Therefore, the proposed method is likely a good solution for protecting audio copyright in an online streaming environment. [ABSTRACT FROM AUTHOR]

Sound event detection (SED) is a widely studied field that has achieved considerable success. The dynamic routing mechanism of capsule networks has been used for SED, but its performance in capturing global information of audio is still limited. In this paper, we propose a method for SED that by combining the capsule network with transformer leverages the strength of transformer in capturing global features with that of capsule network in capturing local features. The proposed method was evaluated on the DCASE 2017 Task 4 weakly labeled dataset. The obtained F-score and Equal Error Rate are 60.6% and 0.75, respectively. Compared to other baseline systems, our method achieves significantly improved performance. Keywords : Sound event detection, audio tagging, gated convolution, transformer, capsule network. [ABSTRACT FROM AUTHOR]

The variational level set model is widely used in image segmentation tasks due to its sound theoretical basis and reliable performance. However, it often faces limitations when dealing with images that contain intensity inhomogeneity and noise. To address this challenge, a novel variational level set model based on local-global function approximation (LSM-LGA) is proposed in this paper. In this approach, an approximation function space is first introduced for the input image. This space defines the function as a linear combination of a local approximation image and a global one, weighted by two coefficients and two complementary characteristic functions defined by a level set function. To achieve the optimal representation of the input image within this approximation function space, a hybrid-metric strategy that combines Euclidean distance and Jeffreys divergence is introduced. An alternating direction method of multiplier (ADMM) based on Euclidean distance is developed to solve for the optimal weight coefficients of the local and global approximation images. Additionally, gradient descent based on Jeffreys divergence is utilized to solve for the optimal level set function. Furthermore, the existence and uniqueness of the optimal approximation function are proven using the theory of projection onto convex sets. Extensive numerical experiments on natural and synthetic images with intensity inhomogeneity, textures, and noise validate that the LSM-LGA exhibits superior performance compared with several state-of-the-art models in terms of segmentation quality and computational efficiency. [ABSTRACT FROM AUTHOR]

Due to the intricate underwater environment, underwater images frequently exhibit issues such as low contrast, detail loss, and color distortion, which significantly affect underwater operations. To alleviate these concerns and improve image quality, multi-scale cross-layer feature interaction GAN for underwater image enhancement model (MCFIGAN) is proposed in this paper. In the proposed model, we have designed a cross-layer feature interaction module (CLFIM) that fosters the fusion and transmission of features at various levels. This approach aims to extract more comprehensive feature information and enhance the robustness of the network model. Additionally, we have constructed a multi-scale feature module (MSFEM) to extract features at different levels and scales, enabling the model to gain a more comprehensive understanding of the object. By utilizing CLFIM and MSFEM, the generator is capable of producing a generated image that contains a wealth of feature information. Then, an enhanced discriminator module (EDM) is designed to improve the model's ability to identify generated images, thereby encouraging the generator network to generate more realistic underwater images. Finally, several training sessions utilizing multiple joint losses yield underwater images with characteristics of high contrast and rich detail. Through simulation experiments conducted on EUVP dataset and UIEB dataset, our proposed model is found to be superior to related models in terms of visual effect and quantitative index, particularly in contrast enhancement, detail recovery, and color restoration. [ABSTRACT FROM AUTHOR]

Due to the capability of supporting high-speed data communication with low energy cost, visible light communication has emerged as a promising supplement to radio frequency communications. However, the inherent nonlinearity caused by the light emitting diode could distort the signal, and thus posing a challenging problem for the system performance. In order to mitigate the nonlinear effect, reproducing kernel Hilbert spaces based schemes are recently investigated, where the minimum symbol error rate (MSER) criterion is proposed as a more suitable criterion compared to the commonly used minimum mean square error (MMSE), since MSER directly optimizes the symbol error rate. In this paper, we present a new post-distortion scheme by exploiting the MSER criterion based kernel affine projection algorithm to combat the signal distortion. In addition, to avoid the linearly increased computational complexity, the surprise criterion is introduced for sparsification. Simulations are carried out to validate the performance, and the numerical results show that the proposed method can obtain remarkable performance improvement over the available algorithms. [ABSTRACT FROM AUTHOR]