Search

As a result of its aggressive nature and late identification at advanced stages, lung cancer is one of the leading causes of cancer-related deaths. Lung cancer early diagnosis is a serious and difficult challenge that is crucial to a person's survival. The first diagnosis of the malignant nodules is typically made using chest radiography (X-rays) and computed tomography (CT) scans; however, the potential presence of benign nodules results in incorrect conclusions. The early phases of both benign and malignant nodules exhibit striking similarities. In this paper, a novel deep learning-based model is proposed for the precise diagnosis of malignant nodules. The proposed approach consists of two stages namely, pre-processing and lung nodule detection. Initially, the Lung CT scan images are collected from the dataset. Then, to remove the noise present in the input image, we apply an adaptive median filter. Then, to enhance the image, Contrast Limited Adaptive Histogram Equalization (CLAHE) is applied. After pre-processing, the image is given to the optimized mask RCNN classifier to detect the malignant and benign nodules. To enhance the performance of the Mask RCNN classifier, the hyper-parameters are optimally selected using hybrid particle snake swarm optimization (PS2OA). The proposed PS2OA is a hybridization of particle swarm optimization (PSO) and snake swarm optimization (SSO). The performance of the proposed approach is analyzed based on different metrics and effectiveness compared with state-of-the-art works. The proposed approach attained the maximum accuracy of 97.67%. This work aimed at assisting radiologists to detect and diagnose small-size pulmonary nodules more accurately. [ABSTRACT FROM AUTHOR]

Currently, security enhancement of biometric systems is an important issue that deserves consideration. This is attributed to the threats facing traditional recognition systems, which depend on Personal Identification Numbers (PINs) that can be stolen, easily. Utilization of original biometrics to access user services may lead to loss of the biometrics forever, if hacking attempts succeed in gaining access to the storage database of original templates. To address this concern and to avoid the utilization of original biometrics, we keep them away from being compromised through the utilization of cancelable biometric templates. This paper introduces a novel methodology for user authentication with multiple biometrics to generate distorted non-invertible cancelable templates to be stored in the database. The proposed framework begins with Discrete Cosine Transform (DCT) to achieve data compression in a multi-biometric scenario. After that, Double Random Phase Encoding (DRPE) is applied to increase the security level of the generated templates. Finally, an adaptive filter is used to induce an effect of whitening to generate the cancelable biometric templates. The generated patterns are uncorrelated due to the effect of encryption and adaptive filtering, which improves the security level against identity theft and provides good performance. Simulation results prove a good performance of the proposed cancelable biometric recognition framework with an Area under the Receiver Operating Characteristic curve (AROC) of 52.12% and an Equal Error Rate (EER) of 44.8462%. [ABSTRACT FROM AUTHOR]

Pneumonia is a serious inflammatory disease that causes lung ulcers, and it is one of the leading reasons for pediatric death in the world. Chest X-rays are perhaps the most commonly utilized modalities to recognize pneumonia. Generally, the illness could be analyzed by a specialist radiologist. But for some reason, the diagnosis may be subjective. Thus, the physicians must be guided by computer-aided diagnosis frameworks in this challenging task. In this study, we propose a combined deep learning architecture to identify pneumonia in chest radiography images. We first, use Adaptive Median Filter for images enhancement, then we employ a regularized Convolutional Neural Network for features extraction, and then we use Long Short Term Memory as a classifier. Finally, the attention mechanism is used to direct the network attention to relevant features. The suggested approach was tested on two publicly available pneumonia X-ray datasets provided by Kermany and the Radiological Society of North America. On the Kermany and RSNA datasets, the suggested technique attained accuracy rates of 99.91% and 88.86%, respectively. In the last stage of our experiments, we employed a Grad-CAM-based color visualization technique to precisely interpret the detection of pneumonia in radiological images. The results outperformed those of state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

The filtered-x least mean square (F-xLMS) technique dependent distributed active noise control (DANC) system is less effective against outliers and disturbances that represent impulses. In such an environment, DANC algorithms based on Swish functions are being developed to reduce noise. The Swish function overcomes the vanishing gradient problem and accelerates convergence by utilising the smoothness and unboundedness properties. In light of this, the current manuscript presents a set of robust techniques for DANC systems that utilizes the Swish structure in lieu of diffusion strategy for filtered-x Blake Zisserman, filtered-x logarithmic hyperbolic cosine adaptive filter algorithms, filtered-x least mean fourth, and filtered-x generalized maximum correntropy criterion. In regard to the suggested methods, weight updating rules pertaining utilizing to the introduced algorithms are created and a convergence analysis is done. The proposed methods achieve quantitatively 1–5 dB enhancement in noise mitigation in case of varied impulsive noise models and input reference signals considered with a faster convergence compared to previous strategies. [ABSTRACT FROM AUTHOR]

Electronic equipment used on a daily basis now frequently includes active noise cancellation. The adaptive filters, which are positioned within, are essential for noise cancellation. An essential component to take into account for the overall performance is the structural and computational complexity of the filter. The filter's structure has an impact on this. The amount of taps determines the structure. Active noise cancellation filters often have set tap lengths and are lengthy, which causes sluggish convergence and delay. As a result, a trade-off between the filter's length and convergence is required. This is conceivable if there is a flexible filter with a tap length that adapts to the environment while still ensuring acceptable convergence. This study proposes a novel Minimum Mean M-estimate method with changeable tap length and uses a sub-band adaptive filtering technique to shorten the filter's length. In order to maximize the filter's efficiency, the advantages of three approaches are specifically merged in this work. They are the proposed algorithm, the proposed method's variable tap length variant, and the sub-band adaptive filtering. The simulation's findings and recommendations are supported. [ABSTRACT FROM AUTHOR]

Low-light image enhancement algorithms have been widely developed. Nevertheless, using long exposure under low-light conditions will lead to motion blurs of the captured images, which presents a challenge to address low-light enhancement and deblurring jointly. A recent effort called LEDNet addresses these issues by designing a encoder-decoder pipeline. However, LEDNet relies on paired data during training, but capturing low-blur and normal-sharp images of the same visual scene simultaneously is challenging. To overcome these challenges, we propose a self-supervised normalizing flow called SSFlow for jointing low-light enhancement and deblurring. SSFlow consists of two modules: an orthogonal channel attention U-Net (OAtt-UNet) module for extracting features, and a normalizing flow for correcting color and denoising (CCD flow). During the training of the SSFlow, the two modules are connected to each other by a color map. Concretely, OAtt-UNet module is a variant of U-Net consisting of an encoder and a decoder. OAtt-UNet module takes a low-light blurry image as input, and incorporates an orthogonal channel attention block into the encoder to improve the representation ability of the overall network. The filter adaptive convolutional layer is integrated into the decoder, applying a dynamic convolution filter to each element of the feature for effective deblurring. To extract color information and denoise, the CCD flow makes full use of the powerful learning ability of the normalizing flow. We construct an unsupervised loss function, continuously optimizing the network by using the consistent color map between the two modules in the color space. The effectiveness of our proposed network is demonstrated through both qualitative and quantitative experiments. Code is available at https://github.com/shibaoshun/SSFlow. [ABSTRACT FROM AUTHOR]

Nonlinear ocean waves have a significant impact on the functioning of several offshore activities. Predicting the internal ocean waves plays a crucial role on submarine and ship operations. Data assimilation is a mechanism in which data observed is interpreted, processed and adapted. The existing works for estimating the future atmospheric condition are highly dependent on the exact initial state, which mostly differ from the observation. This paper proposes modelling of internal ocean waves using automatic internal wave detection and data assimilation. Ensemble Kalman filtering method is used to model ocean waves. The proposed system is focused on satellite images. The images are pre-processed for speckle noise using adaptive filters. Enhanced residual network is used for edge detection. Unlike the existing edge detection methods that have high complexity, this enhanced residual network works with low complexity and makes a direct mapping between the input wave image and wave edge. Finally, the potential edges of the internal wave are detected and adapted using ensemble Kalman filter. Adaptive thresholding technique is used to determine the appropriate threshold to segregate objects from background. The proposed enhanced edge detection model is compared w.r.t to the parameters weighted cross-entropy loss function, accuracy and root mean squared error with canny edge detection and proved to be better. The detection of internal wave is demonstrated, and the accuracy of the approach is 91% with low RMSE when compared to existing works. [ABSTRACT FROM AUTHOR]

Temporal adaptive convolution has demonstrated superior performance over static convolution techniques in video understanding. However, it needs to be improved in long-time series modeling and multi-scale feature-map adaptation. To address these challenges, we introduce spatio-temporal hybrid adaptive convolution (STHAC), designed to enhance the spatio-temporal modeling capabilities of convolution. This is achieved by learning a set of spatio-temporal calibration filters to mitigate the spatial invariance intrinsic to static convolution methods. Specifically, STHAC learns a linear combination of N adaptive filters by parallelizing two lightweight attention branches. The resulting linearly mixed filters incorporate spatial multi-scale prior knowledge and long-range temporal dependencies. These spatio-temporal calibration filters modulate each frame's static convolutional weight parameters, thereby endowing static convolution with spatial multi-scale adaptability and long-range temporal modeling capabilities. Compared to other dynamic convolution methods, our proposed calibration filters require fewer parameters and incur lower computational complexity. Moreover, we introduce an Omni-dimensional aggregation module to augment the spatio-temporal modeling capacity of STHAC. When combined with STHAC, this aggregation module forms the spatio-temporal adaptive module (STAM) that can replace static convolution. We implement a spatio-temporal dynamic network based on STAM to validate our approach. Experimental results indicate that our model is competitive with state-of-the-art convolutional neural network architectures on action recognition benchmarks such as Kinetics-400(K400) and Something-Something V2(SSV2). [ABSTRACT FROM AUTHOR]

As image–text matching (a critical task in the field of computer vision) links cross-modal data, it has captured extensive attention. Most of the existing methods intended for matching images and texts explore the local similarity levels between images and sentences to align images with texts. Even though this fine-grained approach has remarkable gains, how to further mine the deep semantics between data pairs and focus on the essential semantics in data remains to be quested. In this work, a new semantic filtering and adaptive approach (FAAR) was proposed to ease the above problem. To be specific, the filtered attention (FA) module selectively focuses on typical alignments with the interference of meaningless comparisons eliminated. Next, the adaptive regulator (AR) further adjusts the attention weights of key segments for filtered regions and words. The superiority of our proposed method was validated by a number of qualitative experiments and analyses on the Flickr30K and MSCOCO data sets. [ABSTRACT FROM AUTHOR]

Inexpensive cameras, different acquisition devices and mobile phones have resulted in the production of digital images in all spheres of life. There is common of these images for extracting useful information in many fields to perform image analysis and interpretation. One of the problems being faced in the use of this information is the presence of random valued impulse noise. Impulse noise removal is a crucial step in computer vision. Many articles have been published to remove or suppress impulse noise from grayscale digital images using different approaches. Edges and details are, however, often lost during the restoration of digital images. In this research, a two-stage approach for edge detection, preservation, and impulse noise removal is proposed. An effective statistics rank ordered absolute difference (ROAD) is used in the first stage to detect noisy pixels. Trimmed global mean (TGM) with adaptive window is employed later for the filtering process. With the combination of these processes in the proposed method, a higher quality image is restored and thus becomes more suitable for extracting the desired information. Comparison with many state-of-the-art methodologies indicates that the proposed method is effective in the detection of corrupted pixels and the restoration of images. [ABSTRACT FROM AUTHOR]

In this paper, a novel power management strategy (PMS) is proposed for optimal real-time power distribution between battery and supercapacitor hybrid energy storage system in a DC microgrid. The DC-bus voltage regulation and battery life expansion are the main control objectives. Contrary to the previous works that tried to reduce the battery current magnitude and its fluctuation to increase the battery life, the battery thermal considerations, which are a key factor in battery life, are also considered as an objective function in the proposed PMS. Also, the quick charging capability of the battery is another objective in the optimization problem. To solve this multi-objective problem, the Karush–Kuhn–Tucker conditions have been used to make the PMS suitable for real-time operation. In addition, one of the important inputs of conventional PMS is the amount of grid excess/deficiency power. Due to the use of the adaptive Kalman filter estimator, another superior of the proposed PMS is using local data and no need for communication and knowledge of the load or generation power. Using simulation, the performance of the proposed method is compared with three traditional methods filter-based, droop control, and fuzzy logic in terms of controlling the battery current, battery temperature, and battery health, and its superiority has been shown. Finally, the validation of the proposed PMS is experimentally tested on a real setup. [ABSTRACT FROM AUTHOR]

With regards to the problem of multidimensional signal processing in the field of adaptive filtering, geometric algebra based higher-order statistics algorithms were proposed. For instance, to express a multidimensional signal as a multi-vector, the adaptive filtering algorithms described in this work leverage all of the benefits of GA theory in multidimensional signal processing. GA space is employed to extend the traditional least-mean absolute third (LMAT) and newly deduced least-mean mixed third-fourth (LMMTF) adaptive filtering methods for multi-dimensional signal processing. The objective of the presented GA-based least-mean absolute third (GA-LMAT) and GA-based least-mean mixed third-fourth (GA-LMMTF) algorithms is to minimize the cost functions by using higher-order statistics of the error signal e(n) in GA space. The simulation's results revealed that at significantly smaller step size, the given GA-LMAT algorithm is better than the others in terms of steady-state error and convergence rate. Besides, the defined GA-LMMTF algorithm mitigates for the instability of GA-LMAT as the step size increases and illustrates an improved performance relative to mean absolute error and convergence rate. [ABSTRACT FROM AUTHOR]

The Common-path Coherent-dispersion Spectrometer (CODES), an exoplanet detection instrument, executes high-precision Radial Velocity (RV) inversions by recording the phase shifts of interference fringes. Salt-and-pepper noise caused by factors such as improper operation of the CCD probe/analog-to-digital converter and strong dark currents may interfere with the phase information of the fringe. This lowers the quality of the interfering fringe image and significantly interferes with the RV's inversion. In this study, an adaptive median filtering algorithm (CODESmF) based on submaximum and subminimum values is designed to eliminate the interference fringe image's salt-and-pepper noise as well as to reduce RV error. This allows the interference fringe image's phase information to be retained more completely. The algorithm consists of two major modules. Pixel Sub-extreme-based Filtered Noise Monitoring Module: discriminates signal pixels and noise pixels based on the submaximum and subminimum values of the pixels in the filtering window. Adaptive Median Filter Noise Suppression Module: the signal pixel is kept at the original value output, the noise pixel serves as the filtering window's center pixel, and the adaptive median filtering procedure is repeated numerous times with various filtering window sizes. According to the experimental findings, the CODESmF outperforms comparable algorithms and works better at recovering interference fringes. More than 90% of the phase/RV error caused by salt-and-pepper noise is typically eliminated by the CODESmF algorithm, and in certain circumstances, it can even remove roughly 98% of the phase error. [ABSTRACT FROM AUTHOR]

Spina bifida in the foetus is a neurological condition that develops when the spinal cord fails to close properly during pregnancy. A birth defect that causes permanent paralysis in affected babies. Spina bifida is a birth defect that can be treated well if identified and treated early. It is generally agreed that ultrasonic imaging is the gold standard for foetal monitoring. In this research, ultrasound pictures of the foetal spine were segmented and classified into normal and pathological categories using two distinct deep learning algorithms. Three hundred pregnant women were recruited between November 2015 and November 2020 and subjected to three-dimensional ultrasound exams at a hospital. Using an adaptive bilateral filter (ABF), any visible noise in the pictures is eliminated. Segmentation of foetal spina bifida images is performed using the dilated encoder-decoder (DED) technique, and the images are then classified using a Feature Map-based differential convolutional network (FMDDCN). Analysis of segmentation showed that the suggested model was 96% accurate in terms of pixels and 87% accurate in terms of mean intersection over union (MIoU). At the end of the classification evaluation, the suggested model had an accuracy of 96.5, whereas the state-of-the-art methods only managed 95.0. [ABSTRACT FROM AUTHOR]

Electrocardiogram is the most essential diagnostic test for heart disease detection in this era, where it has low frequency and small amplitude, making it vulnerable to a variety of stimuli, including high/low-frequency noises resulting in diagnostic quality suffers. Implementation of adaptive filter removes noise from the signal in a better way. Adaptive filters in signal processing plays a major role in biomedical applications for denoising different types of noises such as Power Line Interference noise which is generated by the power line electromagnetic field and it exhibits its peak from 50 to 60 Hz, Baseline Wander noise which occurs due to the variation of electrode skin impedance, Motion Artifacts which is generated by the electrode motions away from the contact zone on the skins and Muscle Noise, this is due to the Electromyographic signals, originating from the skeletal muscle contractions. In order to eliminate the above-mentioned noises, the most common adaptive filters such as Least Mean Square (LMS) and Recursive Least Squares (RLS) are used. In existing method, they used Cascaded Least Mean Square adaptive filters to reduce noises at a much higher rate than the Least Mean Square Adaptive filter and to achieve the higher Signal to Noise Ratio (SNR). The Cascaded RLS filter can be used to obtain faster range of convergence than the Cascaded LMS. The filter is designed with adaptive variable step size which calculates the step size according to the input signal of the filter, this is used to increase the stability of the filter. The systolic architecture and associativity of the VLLMS filter is implemented to reduce the number of adders and multipliers along the critical path of the filter. This work comprises of designing the 4 tap, 8 tap and 16 tap cascaded systolic associative variable leaky least mean square adaptive filters using Simulink and the SNR comparisons have been made between normal LMS cascaded filter and cascaded VLLMS filter. The area and delay between normal filter and systolic associative filter are compared by converting the Simulink design into Verilog and by implementing in Xilinx. [ABSTRACT FROM AUTHOR]

Background: Residual image noise is substantial in positron emission tomography (PET) and one of the factors limiting lesion detection, quantification, and overall image quality. Thus, improving noise reduction remains of considerable interest. This is especially true for respiratory-gated PET investigations. The only broadly used approach for noise reduction in PET imaging has been the application of low-pass filters, usually Gaussians, which however leads to loss of spatial resolution and increased partial volume effects affecting detectability of small lesions and quantitative data evaluation. The bilateral filter (BF) — a locally adaptive image filter — allows to reduce image noise while preserving well defined object edges but manual optimization of the filter parameters for a given PET scan can be tedious and time-consuming, hampering its clinical use. In this work we have investigated to what extent a suitable deep learning based approach can resolve this issue by training a suitable network with the target of reproducing the results of manually adjusted case-specific bilateral filtering. Methods: Altogether, 69 respiratory-gated clinical PET/CT scans with three different tracers ( [ 18 F ] FDG, [ 18 F ] L-DOPA, [ 68 Ga ] DOTATATE) were used for the present investigation. Prior to data processing, the gated data sets were split, resulting in a total of 552 single-gate image volumes. For each of these image volumes, four 3D ROIs were delineated: one ROI for image noise assessment and three ROIs for focal uptake (e.g. tumor lesions) measurements at different target/background contrast levels. An automated procedure was used to perform a brute force search of the two-dimensional BF parameter space for each data set to identify the "optimal" filter parameters to generate user-approved ground truth input data consisting of pairs of original and optimally BF filtered images. For reproducing the optimal BF filtering, we employed a modified 3D U-Net CNN incorporating residual learning principle. The network training and evaluation was performed using a 5-fold cross-validation scheme. The influence of filtering on lesion SUV quantification and image noise level was assessed by calculating absolute and fractional differences between the CNN, manual BF, or original (STD) data sets in the previously defined ROIs. Results: The automated procedure used for filter parameter determination chose adequate filter parameters for the majority of the data sets with only 19 patient data sets requiring manual tuning. Evaluation of the focal uptake ROIs revealed that CNN as well as BF based filtering essentially maintain the focal SUV max values of the unfiltered images with a low mean ± SD difference of δ SUV max CNN , STD = (−3.9 ± 5.2)% and δ SUV max BF , STD = (−4.4 ± 5.3)%. Regarding relative performance of CNN versus BF, both methods lead to very similar SUV max values in the vast majority of cases with an overall average difference of δ SUV max CNN , BF = (0.5 ± 4.8)%. Evaluation of the noise properties showed that CNN filtering mostly satisfactorily reproduces the noise level and characteristics of BF with δ Noise CNN , BF = (5.6 ± 10.5)%. No significant tracer dependent differences between CNN and BF were observed. Conclusions: Our results show that a neural network based denoising can reproduce the results of a case by case optimized BF in a fully automated way. Apart from rare cases it led to images of practically identical quality regarding noise level, edge preservation, and signal recovery. We believe such a network might proof especially useful in the context of improved motion correction of respiratory-gated PET studies but could also help to establish BF-equivalent edge-preserving CNN filtering in clinical PET since it obviates time consuming manual BF parameter tuning. [ABSTRACT FROM AUTHOR]

Artificial intelligence assists medical practitioners with disease diagnosis from acute to chronic illnesses and has brought significant breakthrough in image-oriented analysis of diseases. Skin cancer is one of the deadliest diseases, posing a great threat to human health. Skin cancer warrants early and precise examination of skin lesions to prevent mortality. Skin lesions can be examined by biopsy, an invasive procedure to remove piece of tissue, tested in a laboratory. Biopsy is a painful procedure. The predictive models in artificial intelligence are found to be useful in analyzing skin lesions-based images. Hence, in this paper, we have analyzed 10,015 skin lesion-based images and classified seven types of skin diseases among which, one of them is melanoma, a form indicative of skin cancer. This work proposes, a novel APFB (Adaptive Percentage Filger for Binarization) to remove the hairline noise from the skin images. The performance of the APFB filter is compared with the state-of-art hairline noise removal methods such as Contrast Limited Adaptive Histogram Equalization (CLAHE) and Bilateral methods and proven to be best in terms of image quality analysis metrics such as PSNR, SSIM, BRISQUE and NIQE. The original images and hairline noise removed images are then passed into custom built CNN models for skin disease classification. The model examines the lesions features like pigment, texture, color, lines, shapes and symmetry to classify the different types of skin diseases. Among the four CNN models we created, the CNN model that takes images of APFB hair line noise removal method obtains 99.98% training accuracy and 100% testing accuracy than the other models. Thus, the model may also be used to predict skin cancer without a biopsy. [ABSTRACT FROM AUTHOR]

Constrained least mean square (CLMS) algorithm is the most popular constrained adaptive filtering algorithm due to its simple structure and easy implementation. However, its convergence slows down when the input signal is colored. To address this issue, this paper firstly introduces the normalized subband adaptive filter (NSAF) into the constrained filtering problem and derives a constrained NSAF (CNSAF) algorithm using the Lagrange multiplier method. Benefiting from the good decorrelation capability of the NSAF, the proposed CNSAF algorithm significantly improves the convergence performance of the CLMS algorithm under colored inputs. Then, the mean and mean-square stability of the CNSAF algorithm is analyzed, and the theoretical models to characterize the transient and steady-state mean square deviation (MSD) behaviors of the CNSAF algorithm are derived utilizing the Kronecker product property and vectorization method. Further to extend the CNSAF algorithm to the problem of sparse system identification, a sparse version of the CNSAF algorithm (S-CNSAF) is derived. Finally, the validity of the derived theoretical MSD prediction models and the superiority of the proposed algorithms are confirmed by extensive computer simulations on system identification with colored inputs. [ABSTRACT FROM AUTHOR]

Based on the criterion of minimum error entropy, this paper proposes a novel conjugate gradient algorithm, called MEE-CG. This algorithm has robust performance under non-Gaussian interference. Theoretical analysis and experimental results demonstrate that the proposed algorithm displays more robust performance than the conventional conjugate gradient methods on the basis of the mean square error and the maximum correntropy criterion. Compared with the stochastic gradient minimum error entropy algorithm and the recursive minimum error entropy algorithm, the proposed algorithm provides a trade-off between computational complexity and convergence speed. [ABSTRACT FROM AUTHOR]

One of the most relevant tasks of high-dimensional estimation is the computation of the parameters covariance matrix. This matrix offers valuable insights into the uncertainties associated with the estimation process. In the context of adaptive filtering, the weight error covariance matrix is a key parameter that determines how the filter adapts to input statistics and noise levels. Additional statistical information about the weight error vector allows one to depict a more precise description of the stochastic coupling between the adaptive weights. This paper concentrates on an in-depth study of the least mean squares asymptotic weight error covariance matrix, employing the exact expectation analysis. Through this examination, a key conclusion emerges: The symmetries commonly engendered by traditional analyses are not present in the actual covariance matrix. Instead, these symmetries are artifacts of the widespread assumption of independence. In summary, the advanced analysis performed in this work reveals a significantly more nuanced learning behavior exhibited by the least mean squares algorithm, challenging the conventional understanding put forth by traditional approaches. The theoretical predictions are confirmed by extensive simulations. [ABSTRACT FROM AUTHOR]

Gaussian filter (GF) is a commonly used linear filter in signal and image noise reduction. However, its limitation is that it cannot adapt parameters to deal with non-stationary noise that varies over time. To address this problem and improve the filtering effectiveness of GF in the face of non-stationary non-Gaussian (NSNG) noise, this paper proposes a new approach called adaptive Gaussian filter based on improved complete ensemble empirical mode decomposition (ICEEMDAN-AGF). The ICEEMDAN-AGF firstly uses the fusion information of the dispersion entropy (DE) and the power spectral entropy (PSE) to divide the intrinsic mode functions (IMFs) into two groups. One group is called guiding IMFs, which contains the high-frequency components of the NSNG noise, and the other group is called hybrid IMFs, which contains the low-frequency components of the NSNG noise and all the noise-free signals. Next, a method called multi-resolution local similarity (MRLS) is proposed to identify the mixed modes presented in the guiding IMFs. Then, the variance of the guiding IMFs is used to adjust the window width w and kernel parameter σ of GF. Finally, the adaptive Gaussian filter (AGF) obtained above is used to filter the hybrid IMFs. The experiments shows that ICEEMDAN-AGF performs better than other conventional algorithms on known signals. [ABSTRACT FROM AUTHOR]

A new model-free and data-adaptive feature screening procedure referred to as the concordance filter is developed for ultrahigh-dimensional data. The proposed method is based on the concordance filter which measures concordance between random vectors and can work adaptively with several types of predictors and response variables. We apply the concordance filter to deal with feature screening problems emerging from a wide range of real applications, such as nonparametric regression and survival analysis, among others. It is shown that the concordance filter enjoys the sure screening and rank consistency properties under weak regularity conditions. In particular, the concordance filter can still be powerful in the presence of censoring and heavy tails. We further demonstrate the superior performance of the concordance filter over existing screening methods by numerical examples and medical applications. [ABSTRACT FROM AUTHOR]

Distributed inference tasks could be performed by adaptive filtering techniques. Several enhancement strategies for such techniques were proposed, such as sparsity-aware algorithms, coefficients reuse and correntropy-based cost functions in the case of impulsive noise. In this paper, a general framework based on Lagrange multipliers for the derivation of sophisticated algorithms that incorporate most of these improvements is described. A new general identification algorithm is derived as an example of the proposed approach and its performance is assessed in a distributed setting. [ABSTRACT FROM AUTHOR]

Global Navigation Satellite System (GNSS) can provide an approach for spacecraft autonomous navigation in earth–moon space to make up for the insufficiency of earth-based tracking, telemetry, and control systems. However, its weak power and poor observation geometry near the moon causes new problems. After the GNSS signal characteristics and satellite visibility were evaluated in Phasing Orbit and Lunar Transfer Orbit, we proposed an adaptive Kalman filter based on the Carrier-to-Noise ratio (C/N0) and innovation vector to weaken the influence of GNSS accuracy attenuation as much as possible. The experimental results show that the spacecraft position and velocity accuracy are better than 10 m and 0.1 m/s near the Earth, and better than 50 m and approximately 0.2 m/s near the moon use GNSS with the proposed adaptive algorithms. Additionally, because of the deterioration of navigation performance based on the orbit filter during orbital maneuvering, we used accelerometer data to compensate for the dynamic model to maintain navigation performance. The results of the experiment provide a reference for subsequent studies. [ABSTRACT FROM AUTHOR]

Powerline interference (PLI) more frequently affects the electrocardiogram (ECG) signal during the recording and acquisition. This paper proposes a unique method for suppressing the PLI in the ECG signal using an adaptive filter (AF) with a modified lightning search algorithm (MLSA). The MLSA approach automatically updates the filter coefficient weights to minimize the error between the noisy ECG signal and the filter output. The input signal of a filter is also optimized based on the error signal using MLSA. The proposed methodology's effectiveness is verified by adding PLI noise to the ECG records from the MIT-BIH arrhythmia database at different signal-to-noise ratios (SNRs). The potency of the proposed methodology is assessed by evaluation metrics, namely SNR improvement, mean square error (MSE), mean absolute error (MAE), percentage of root-mean-square difference (PRD), Huber loss (HL), and the correlation coefficient (CC). The results of the proposed methods outperform the existing techniques of empirical mode decomposition, variational mode decomposition with a non-local mean approach, empirical wavelet transform, variational mode decomposition with a variable notch filter, and an AF with the least mean square (LMS) approach in terms of evaluation metrics and visual observation. [ABSTRACT FROM AUTHOR]

To address the shortcomings of the widely-linear complex-valued distributed adaptive filtering algorithm that cannot combine convergence speed and steady-state performance when using a fixed step-size strategy. In this paper, by making the partial derivative of the square of the sum of a priori error and a posteriori error with respect to step-size equal to zero, we propose a novel variable step-size (VSS) strategy, which is formulated as a function of a priori error and a posteriori error. In addition, a corresponding adaptive filtering algorithm is proposed by using the VSS strategy. Then in the process of parameter update, the moving average method is used to avoid large perturbations in iteration of the algorithm. Secondly, the theoretically analysis of the transient and steady-state performance of the proposed algorithm is provided. Finally, in the numerical simulation experiments, the effect of each parameter on the algorithm is tested, the result of the comparison with the fixed-step algorithm illustrates that the proposed algorithm has more superiority, and the high degree of matching between theory and experiment also verifies the feasibility and accuracy of the theoretical analysis method. [ABSTRACT FROM AUTHOR]

While the practical application of adaptive filters has indeed garnered substantial attention, two pressing issues persist that have a profound impact on their performance—system sparsity and the presence of contaminated Gaussian impulsive noise. In this research paper, we propose a novel approach to tackle both of these issues simultaneously by introducing the concept of a proportionate matrix. Specifically, we present a proportionate maximum total complex correntropy algorithm based on the errors-in-variables model. The paper presents a theoretical analysis of the steady-state weight error power under the influence of impulsive noise. Furthermore, it discusses the performance comparison in system identification and highlights the robustness of the proposed algorithm. To validate its effectiveness, a simulation involving stereophonic acoustic echo cancellation is conducted, and the results confirm the clear advantages of the proposed Proportionate Maximum Total Complex Correntropy algorithm. [ABSTRACT FROM AUTHOR]

Time–frequency analysis methods represent an efficacious approach to process data derived from full-waveform acoustic logging. This research proposes a coupled time–frequency analysis method that synergistically incorporates ensemble empirical mode decomposition, fractional Fourier transform, Choi–Williams (CW) distribution and adaptive filtering. This coupled method facilitates the comprehensive analysis of full-waveform acoustic logging signals within rigid geological formations. Specifically, it is designed to explore the time–frequency characteristics of shear wave (S-wave) within the context of igneous reservoirs. The coupled time–frequency analysis method significantly ameliorates the time–frequency coupling phenomenon of full-waveform acoustic logging signal. This coupled method is applied across different wells to confirm its wide-ranging applicability and study the time–frequency characteristics response of S-wave through the information of the peak time of the S-wave, the main frequency and the amplitude of the CW distribution. [ABSTRACT FROM AUTHOR]

In the initial alignment of shaking base, Kalman filter is easy to diverge, and the alignment result is poor. Many scholars introduce various nonlinear filtering methods to solve this problem, but these filtering methods not only have large amount of calculation, but also are low accuracy. In theory, standard Kalman filtering can obtain the optimal estimate of the state under the condition that the structural parameters and noise statistical characteristic parameters of the random system are accurately known. Based on the problem, an improved adaptive Kalman filter (IAKF) algorithm is proposed to complete the initial alignment of the shaking base. While performing state estimation, it can also estimate the system's noise parameters in real time through measurement output. The experimental results show that the improved algorithm has high anti-interference and anti-divergence ability. Under the simulation conditions, pitch, roll and heading errors can be stabilized at about 10 degrees. The real measurement with Redmi K30Pro and iPhone 14 mobile phone shows that the alignment result of the improved algorithm is consistent, and the difference of multiple alignment results is stable within 5 degrees, while the difference of traditional method is more than 20 degrees. It can be applied in smart phone navigation with the strapdown inertial navigation system. [ABSTRACT FROM AUTHOR]

A cavity ring-down spectroscopy (CRDS) system based on the time-division multiplexing (TDM) and an improved Whittaker-Henderson (W-H) filtering algorithm was proposed to synchronous detect of methane (CH4) and carbon dioxide (CO2) gas absorption spectroscopy. The mechanical optical switch was used to quickly switch distributed feedback (DFB) lasers with different wavelengths in the optical path.And the adaptive W-H filtering algorithm combined with convolutional neural network (CNN) could quickly and accurately obtain the best weight parameters to achieve effective denoising of absorption spectrum.In addition, the gradient concentration of CH4 and CO2weredetected and the limit of detection (LOD) of CRDS system was studied.The results of experiment demonstrated that the system could detect CO2 and CH4 in real time and had good stability. Allan deviation analysis shows that the LOD of CH4 and CO2 are 8 ppb and 0.85 ppm under the average number of 1. The LOD of CH4 and CO2 can be optimized to 2 ppb and 0.16 ppm under the optimal average number. In short, the proposed system had high sensitivity, good stability and could measure various gases synchronously, which had important application potential in the field of gas monitoring. [ABSTRACT FROM AUTHOR]

In the past decade, the augmented complex-valued least mean square adaptive filtering algorithm based on the minimum mean square error criterion and widely-linear model has attracted much attentions due to its simplicity and applicability to non-circular signals. However, the system noise with impulse characteristics will dramatically affect the convergence of the augmented complex-valued least mean square algorithm. Recently, a robust widely-linear complex-valued least mean M-estimate algorithm has been proposed, which significantly improves the robustness of the augmented complex-valued least mean square algorithm to impulsive noise. But it should be noted that the widely-linear complex-valued least mean M-estimate algorithm with fixed step size faces the trade-off between convergence speed and steady-state accuracy. To overcome this drawback, a variable step size widely-linear complex-valued least mean M-estimate algorithm is presented in this paper, in which the optimal step size at each iteration is obtained by maximizing the difference between the mean square deviations of the WL-CLMM algorithm at adjacent moments. In order to reveal the statistical convergence behavior of the proposed variable-step-size widely-linear complex-valued least mean M-estimate algorithm to better guide the parameter selection in practical applications. Theoretical transient and steady-state mean-square deviation convergence behaviors of the variable-step-size widely-linear complex-valued least mean M-estimate algorithm are analyzed and the corresponding mean-square deviation expressions are also derived. Computer simulations on system identification and stereophonic acoustic echo cancellation in impulsive noise environments confirm the validity of the analysis results and the performance improvement from the variable step size strategy. [ABSTRACT FROM AUTHOR]

The power quality of the grid-coupled photovoltaic system (GCPVS) is enhanced using a control scheme developed by employing a robust generalized soft-root-sign (GSRS) adaptive filter. Power quality (PQ) deterioration issues are common in the distribution grid due to the mix of irregular loads. The GSRS control scheme mitigates harmonics, compensates load unbalances, achieves unity power factor at the grid, and compensates load reactive power, thus leading to balanced sinusoidal currents at the grid side. The GSRS algorithm utilizes a soft-root-sign cost function to achieve less steady-state misalignment while estimating magnitudes of fundamental frequency load current components for PQ improvement. The designed single-stage GCPVS performs smooth power flow between local loads and the grid without necessitating an energy storage system. The control scheme initially meets load demand and then sends any surplus power to the grid. However, if the generated PV power falls below the load demand due to environmental conditions, the additional power for the load is drawn from the grid. Moreover, voltage source converter used in the GCPVS also provides the functionality of DSTATCOM, thus avoiding the usage of additional equipment. The proposed GSRS control scheme and the GCPVS are modeled in MATLAB/Simulink. A laboratory-designed experimental setup is used to carry out the experimental validation of the GSRS control scheme and found to be performing satisfactorily. The behavior of GCPVS is verified during steady-state and dynamic conditions. The GCPVS performance is verified with the standard IEEE-519-2022, and a comparison of the GSRS algorithm with other filtering algorithms is provided. [ABSTRACT FROM AUTHOR]

Existing methods for filter pruning mostly rely on specific data-driven paradigms but lack the interpretability. Besides, these approaches usually assign layer-wise compression ratios automatically only under given FLOPs by neural architecture search algorithms or just manually, which are short of efficiency. In this paper, we propose a novel interpretable task-inspired adaptive filter pruning method for neural networks to solve the above problems. First, we treat filters as semantic detectors and develop the task-inspired importance criteria by evaluating correlations between input tasks and feature maps, and observing the information flow through filters between adjacent layers. Second, we refer to the human neurobiological mechanism for the better interpretability, where the retained first layer filters act as individual information receivers. Third, inspired by the phenomenon that each filter has a deterministic impact on FLOPs and network parameters, we provide an efficient adaptive compression ratio allocation strategy based on differentiable pruning approximation under multiple budget constraints, as well as considering the performance objective. The proposed method is validated with extensive experiments on the state-of-the-art neural networks, which significantly outperforms all the existing filter pruning methods and achieves the best trade-off between neural network compression and task performance. With ResNet-50 on ImageNet, our approach reduces 75.49% parameters and 70.90% FLOPs, only suffering from 2.31% performance degradation. [ABSTRACT FROM AUTHOR]

The quality of yellow peaches was severely reduced due to bruising. Detection of bruised areas in yellow peach is easily disturbed by the calyx and stem ends. Therefore, it is proposed to utilize hyperspectral imaging in conjunction with I-Otsu in the detection of bruise detection in yellow peach after distinguishing between calyx and stem end using curvature-assisted Hough transform circle detection. The band ratio was used to improve the contrast of the images, and the improved Otsu method was used to improve the segmentation accuracy. Noise in the image is eliminated by adaptive median filtering. The effects of the calyx and stem ends are eliminated by curvature-assisted Hough transform circle detection. Spectral bands with valid feature information were selected by principal component analysis and key single-wavelength images (452.8 nm, 608.9 nm, 671.8 nm, 689.4 nm, 825.7 nm, and 966.2 nm) were selected from the loading curves of the spectral regions to create PC images and band ratio images. Band ratio (Q608.9/689.4) images with I-Otsu were used to segment the bruise region. Ultimately, 96% of the bruised yellow peaches were correctly identified. This study demonstrates that hyperspectral imaging combined curvature-assisted Hough transform circle detection and I-Otsu can accurately identify bruised areas as well as calyx and stem ends in yellow peach. [ABSTRACT FROM AUTHOR]

This paper provides the parallel distributed Kalman filter (PDKF) with non-aligned time indexes, which uses four processing nodes to run the linear Kalman filter in parallel. To provide outputs that are in sync with the time indexes that are now in effect, each processing node aggregates prior time indexes utilizing non-aligned time indexes. This method increases system efficiency by enabling simultaneous data sharing among all nodes and parallel processing with earlier time indexes. Moreover, the consistent acknowledgment of each block's processing time by all nodes ensures smooth functioning. With the help of the suggested PDKF, computing-intensive processes can be carried out in a distributed fashion, improving computation efficiency without sacrificing estimation accuracy. Simulation results show that while achieving nearly equivalent convergence performance to the sequentially operated Kalman filter, the proposed PDKF architecture outperforms both the sequential recursive least squares (RLS) and parallel distributed adaptive signal processing RLS algorithms. This study highlights the algorithm's strong overall performance through a thorough discussion, revealing its capacity to lower computing costs without sacrificing estimation accuracy inside the network. The book delves deeper into the algorithm's overall performance, emphasizing its effectiveness beyond simple estimating accuracy. [ABSTRACT FROM AUTHOR]

Knowledge of remaining battery charge is fundamental to electric vehicle deployment. Accurate measurements of state-of-charge (SOC) cannot be obtained directly and estimation methods must be used instead. This requires both a good model of a battery and a well-designed state estimator. Here, hysteretic reduced-order battery models and adaptive extended Kalman filter estimators are shown to be highly effective, accurate predictors of SOC. A battery model parameterisation framework is proposed, which enhances standardised methods to capture hysteresis effects. The hysteretic model is parameterised for three independent NMC811 lithium-ion cells and is shown to reduce voltage RMS error by 50% across 18 h automotive drive-cycles. Parameterised models are used alongside an extended Kalman filter, which demonstrates the value of adaptive filter parameterisation schemes. When used alongside an extended Kalman filter, adaptive covariance matrices yield highly accurate SOC estimates, reducing SOC estimation error by 85%, compared to the industry standard battery model. [ABSTRACT FROM AUTHOR]

Diabetic Retinopathy (DR) is one of the foremost reasons for poor eyesight in the modern globe. An earlier detection of DR is vital in offering effectual treatment. Moreover, detecting the severity levels of DR, such as Microaneurysms (MA), Hemorrhages (HMs), Exudates (EXs) and extra development of blood vessel recognition using fundus images are challenging chores owing to the complex structures and shapes of lesions in the fundus image. This paper aims to develop a technique for DR detection at an early stage. MA is the first symptom of DR that leads to blood leakage in the retina. Here, Fractional Hunter Osprey Optimization-Deep Convolutional Neural Network (FHOO-DCNN) is introduced for MA detection. An input image pre-processing is executed by an adaptive wiener filter and then, optic disc (OD) detection is accomplished. The Bayesian U-Net is used for OD detection, which is tuned by Hunter Osprey Optimization (HOO). The HOO is modelled by combining Hunter–Prey Optimizer (HPO) with the Osprey Optimization Algorithm (OOA). Blood vessel segmentation is conducted utilizing morphological Top-Hat transform. Thereafter, features from the input image, blood vessel segmented image, and OD-detected image are extracted. At last, MA detection is performed by DCNN that is tuned using FHOO. Furthermore, FHOO is joining of Fractional Calculus (FC) concept with HOO. In addition, FHOO-DCNN has acquired high accuracy, sensitivity and specificity of 91.1%, 91.2% and 90.4%. The proposed method is applicable in primary screening and regular clinical work for the monitoring of the progression of DR. [ABSTRACT FROM AUTHOR]

Recursive Least Square (RLS) is a popular algorithm for noise cancellation in non-stationary signals; however, it demands more computational resources and more difficult mathematical operations. Also, RLS has less performance stability. The present research explores a novel idea of an RLS adaptive noise cancellation through a modified method that uses an RLS adaptive filter by introducing an additional constant multiplier, along with their in-depth analysis. The proposed algorithm is analyzed using three primary performance metrics: mean square error (MSE), signal-to-noise ratio (SNR), and convergence rate. The obtained results demonstrate that the proposed algorithm has reduced the MSE by 79.65%, which leads to an improvement in SNR by 86.16% compared to the traditional RLS algorithm. The additional constant multiplier is optimized for SNR, and the optimized value is found to be equal to 0.65, which gives the best possible SNR value of 19.38 dB. Also, the proposed algorithm has been successfully applied to a real-world scenario in acoustic echo cancellation (AEC). The experimental setup for the echo canceller is simulated on MATLAB to measure the echo canceller efficiency in terms of MSE and echo return loss enhancement. Based on performance evaluation, the proposed algorithm has been found to better echo cancellation in AEC as compared to traditional RLS. [ABSTRACT FROM AUTHOR]

This paper proposes a digital morphological filtering method based on a novel structuring element (SE) formulated using fractional Fourier transform (FrFT) and cross-convolution of window functions. The highlighting feature of this newly formulated filter is its flexibility in adding an adaptive nature to classical morphological filtering. Until now, every method listed in the literature makes a convenient assumption that noises corrupting ECG signal are Gaussian and model the filter around this assumption. Addressing this shortcoming, the first-of-its-kind filter adopts the α -stable distribution model (of which Gaussian distribution is a special case) of noise to better replicate the real-time noises interfering with ECG signals. The designed filter can suppress the noise and adapt to the changes in ECG signal morphology for better reconstruction. The proposed filter is tested on MIT-BIH Arrhythmia Database. The simulation results show improved performance in several quantitative metrics, demonstrating the superiority of our suggested method over the currently used state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Visible light communication (VLC) can serve as a supplement to the radio frequency to address its inherent constraints. VLC can be easily implemented in vehicles using their existing LED components. Sunlight interference can cause significant variation in the background noise powers, leading to signal-to-noise ratio (SNR) degradation in vehicular VLC (VVLC) communication. Hence, it is necessary to improve the performance besides interference. Allan variance is a temporal analysis approach that can extract the noise from the received data. In this article, an Allan variance-based adaptive noise canceller for a VVLC system is proposed and evaluated. Based on the Allan variance obtained, the step size and the number of taps of the adaptive filter are adjusted for superior performance. This proposed filter effectively mitigates the pronounced autocorrelation induced by the noise interference. It yields about an 85% bit error rate (BER) reduction compared to the conventional filter. This improved convergence underscores the potential for enhanced communication performance and reliability. The balanced distribution of the Manchester encoded signal outperforms NRZ encoding. Thus, this work contributes a novel adaptive noise canceller design for a VVLC system and also provides valuable insights to the significance of Allan variance in the VLC system performance. [ABSTRACT FROM AUTHOR]

The most common non-invasive diagnostic model is the Electrocardiogram (ECG), which records the heart's electrical activity over time and is used to diagnose various heart conditions. Due to the requirements of a typical eHealth system, it is necessary to compress ECG signals for long-term data recording and remote transmission. Moreover, cardiovascular diseases (CVDs) have been considered the most long-lasting disorders in recent years. The transmission of information from the patient to the distant hospital is necessary because rapid analysis and treatment are essential for the condition to be cured. Also, the data must be in the form of lossless and high-predictability data. So, the goal of this study was to create a two-stage lossless Integer Adaptive Predictor (IAP) compressor that could be implemented on a Field Programmable Gate Array (FPGA) without introducing any data loss during the compression process. Before compression, the ECG signals are denoised using a Fast Normalized Least Mean Square (FNLMS) algorithm-based adaptive filter, which removes the undesirable noise presented in the signal. Here, the adaptive filter is designed based on the hybrid systolic folding structure and compressor-based multiplier architecture to minimize the power, delay and area consumption of the filter while performing the signal-denoising process. Xilinx and MATLAB are used to run simulations using the MIT-BIH Arrhythmia and PTB diagnostic databases. Several performance parameters are used to assess the proposed design's efficacy, and the results are compared to those of similar current designs. Consequently, the proposed compressor achieves a 45.23% compression ratio (CR) on MIT-BIH and a 10.87% average CR on the PTB diagnostic database, which demonstrates that the compression proficiency of the proposed design is high. [ABSTRACT FROM AUTHOR]

To improve the convergence speed and noise cancellation performance of the Delayed Filtered-x Least Mean Square (DFxLMS) algorithm in the hardware implementation of compressor noise cancellation system, an improved Delayed Filtered-x Least Mean Square algorithm based on field-programmable gate array (FPGA) is proposed. The algorithm uses a parallel finite impulse response (FIR) filter structure and DSP48E1 module to implement the parallel multiplication and addition operation, which decreases hardware resource consumption and increases the system's processing speed. The cut set retiming technique reallocates the location of delay units to streamline the algorithm structure, reduce the critical path delay and increase the system data throughput. Moreover, through the adaptive filtering of input and feedback signals of the filter module, the noise suppression effect of the algorithm is improved. The results show that the maximum noise cancellation of the algorithm is 10 dB in the compressor noise cancellation system, which is 3 dB higher than the DFxLMS algorithm, and has 1.5 times higher data throughput. In addition, the Lookup Table (LUT), Flip-Flop (FF), and power consumption decreased by 18.5%, 33.6%, and 26.6%, respectively, and converged 1.8 times as fast. [ABSTRACT FROM AUTHOR]

In this paper, a stochastic model is presented for the nonparametric variable step-size normalized least-mean-square (NP-VSS-NLMS) algorithm. This algorithm has demonstrated potential in practical applications and hence a deeper understanding of its behavior becomes crucial. In this context, model expressions are obtained for characterizing the algorithm behavior in the transient phase as well as in the steady state, considering a system identification problem and Gaussian input data. Such expressions reveal interesting algorithm characteristics that are useful for establishing design guidelines and for the advancement of more refined algorithms. Simulation results for various operating scenarios ratified both the model's accuracy and the algorithm's superior performance relative to other recent and relevant algorithms from the literature. [ABSTRACT FROM AUTHOR]

The study was aimed at developing a new automatic search technique for specific invariant patterns of movement-related brain potentials reflected in multidimensional electroencephalogram (EEG) signals. An adaptive band-pass filter with bandwidth closely matching the spectrum of the desired EEG pattern at the observed moment was synthesized based on the Singular Spectrum Analysis methodology. The preliminary filtering of the original EEG signals provides the required sensitivity for subsequent searching of time boundaries in patterns. The correctness of the developed method was confirmed with standard machine learning tools through the validation of the adaptive search method carried out on the general set of initial data. It is shown that the synthesized method has provided a reliable automatic search for induced pre-movement EEG patterns and the correct determination of their time boundaries (accuracy up 29% on average and reached maximum values to 100% for some individuals). The developed method expands the existing tools to improve the functionality and reliability of various Brain-computer interfaces for various purposes, including medical applications for paralyzed patients. [ABSTRACT FROM AUTHOR]

Within vertebrates, central pattern generators drive rhythmical behaviours, such as locomotion and ventilation. Their pattern generation is also influenced by sensory input and various forms of neuromodulation. These capabilities arose early in vertebrate evolution, preceding the evolution of the cerebellum in jawed vertebrates. This later evolution of the cerebellum is suggestive of subsumption architecture that adds functionality to a pre-existing network. From a central-pattern-generator perspective, what additional functionality might the cerebellum provide? The suggestion is that the adaptive filter capabilities of the cerebellum may be able to use error learning to appropriately repurpose pattern output. Examples may include head and eye stabilization during locomotion, song learning, and context-dependent alternation between learnt motor-control sequences. [ABSTRACT FROM AUTHOR]

A novel multi-combined-step-size normalized subband adaptive filtering algorithm is proposed. Different from the traditional combined-step-size method, by employing a variable mixing parameter on each subband, our designed algorithm is able to combine a large step size with a small one more effectively for each subband. The subband mixing parameters are derived from the variance of the noise-free a priori subband error signals. In this design, a noniterative shrinkage strategy is also utilized to estimate the noise-free a priori subband error signals. Moreover, the mean-square and steady-state performances of the proposed algorithm are studied. Simulation results illustrate that the proposed algorithm outperforms other algorithms mentioned in this paper in terms of tracking capability, convergence speed and steady-state error. [ABSTRACT FROM AUTHOR]

To enhance the performance of the diffusion robust adaptive filtering algorithm, this paper presents a novel loss function framework named the "Tanh cost function framework." By utilizing this framework, we are able to incorporate standard cost functions, devise novel cost functions, and introduce the associated diffusion algorithm. Moreover, the convergence of the algorithm is thoroughly analyzed. Furthermore, the proposed diffusion Tanh algorithm family is simulated in a system identification model to assess its performance, demonstrating superior performance compared to the standard diffusive algorithm. Notably, the DTHLMS within the diffusion tanh algorithm family exhibits superior performance over the mainstream robust diffusion algorithms. [ABSTRACT FROM AUTHOR]

In this paper, an improved high-speed adaptive filter is proposed and implemented using a field-programmable gate array platform. Specifically, a new filter structure combining systolic and convex architectures has been analyzed and compared with conventional filter architectures. The new filter structure efficiently removes power line interference noise from electrocardiogram signals at high convergence speeds. The systolic architecture is used to improve the convergence speed of the filter, and the convex architecture is used in combination to improve the signal-to-noise ratio of the filter. One fast filter was designed using the retimed delay recursive least square algorithm (recursive least square filter design with systolic architecture), and it was combined with one slow filter designed using the least mean square algorithm based on convex combination architecture. The proposed filter architectures are assessed for electrocardiogram noise cancellation, obtained from the MIT-BIH database, and the performance is compared with various filter structures in terms of the signal-to-noise ratio, convergence speed, learning behaviors, and complexity. The results show an improvement in signal-to-noise ratio of 24.2% and an increase in convergence speed of 50% when compared with conventional filter structures. [ABSTRACT FROM AUTHOR]

This paper introduces a reconfigurable AI-enabled scalable median filter with an adaptive impulse detector designed for FPGA-based real-time imaging systems. Its primary objective is to address the degradation of image quality caused by mixed impulsive noise during real-time image transmission and reception. Existing median filters often struggle to provide real-time image processing results that meet high standards in terms of both accuracy and speed. This approach effectively suppresses noise in real-time images while preserving essential edge details, which are crucial for the performance of real-time imaging systems. The algorithm introduces a novel technique of replacing noisy pixels with the processed central value within the image filtering window. This ensures fidelity to the original pixel, which is vital for applications such as image filter cores. To handle high noise densities in real-time systems, the methodology employs a scalable sorting approach for median filtering and an impulse detector, ensuring robust noise reduction without excessive computational complexity. The AI-enabled scalable median filter system achieves a significant reduction in dynamic power consumption, realizing an impressive 46% decrease in power consumption and an 82% reduction in area compared to the existing system. This is particularly beneficial for addressing resource and power-aware constraints in real-time systems. Comprehensive performance evaluation, including metrics such as PSNR, MSE, IEF, and SSIM, demonstrates the efficacy of the filter in enhancing image quality, a critical factor for the success of real-time imaging systems. [ABSTRACT FROM AUTHOR]

Variable digital filter (VDF) plays a significant role in communication and signal processing field. Any prototype filter's preferred frequency response is attained by creating All Pass Transformation (APT) based filter to maintain complete control over the cut-off frequency. However, the speed, power, and area usage of the digital filter are constrained by its performance. Therefore, in this manuscript, All Pass Transformation based Variable digital filters (APT-VDF) using Error Reduced Carry Prediction Approximate Adder (ERCPAA) andSandpiper Optimization fostered Approximate Multiplier (SO-AM) is proposed. The proposed APT-VDF-ERCPAA-SOAM filter design is utilized for enhancing the filter efficiency by reducing noise in the sensor nodes. The proposed ERCPAA design is incorporated with carry prediction and constant truncation for diminishing the path delay and area utilization. Moreover, the proposed SO-AM is used for minimizing the design complexity and power utilization. The simulation of the proposed method is activated in Verilog and the design is synthesized in FPGA uses Xilinx ISE 14.5. The proposed APT-VDF- ERCPAA- SO-AM filter design has attained 35.6%, 21.75%, 28.69% lower power and 46.58%, 12.3%, 38.07% lower delay than the existing approaches, like Very Large-Scale Integration design of All Pass Transformation based Variable digital filters uses a new variable block sized ternary adder (VBSTA) and ternary multiplier (APTVDF-VBSTA-TM), Finite Impulse Response (FIR) adaptive filter design by hybridizing canonical signed digit (CSD) and approximate booth recode (ABR) algorithm in DA architecture (FIR- CSDABR-DA) and digital FIR filter design using Carry Save Adder (CSA) and Structured Tree Multiplier (FIR-CSA-STM) respectively. [ABSTRACT FROM AUTHOR]