Signal ProcessingPub Date : 2025-04-24DOI: 10.1016/j.sigpro.2025.110055
Tao Deng , Lu Lu , Tao Lei , Badong Chen
{"title":"Fixed-point fully adaptive interpolated Volterra filter under recursive maximum correntropy","authors":"Tao Deng , Lu Lu , Tao Lei , Badong Chen","doi":"10.1016/j.sigpro.2025.110055","DOIUrl":"10.1016/j.sigpro.2025.110055","url":null,"abstract":"<div><div>The second-order Volterra (SOV) filter demonstrates excellent performance for modeling nonlinear systems. The main disadvantage of the adaptive SOV filter is that the number of coefficients increases exponentially with memory length, which hinders its practical applications. To circumvent this problem, the sparse-interpolated Volterra filter has been developed. However, the existing algorithms only investigated the performance of gradient-based interpolators and their performance may degrade for combating impulsive noise. A novel fixed-point fully adaptive interpolated Volterra filter under recursive maximum correntropy (FPFAIV-RMC) algorithm is proposed. In particular, the coefficients of the sparse SOV filter are adapted by the RMC algorithm and the coefficients of the interpolator are updated by the fixed-point algorithm under RMC. Additionally, the convergence of the FPFAIV-RMC algorithm is analyzed. The efficacy of the FPFAIV-RMC algorithm is validated by simulations for nonlinear system identification, nonlinear acoustic echo cancellation (NLAEC), and prediction in impulsive noise.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110055"},"PeriodicalIF":3.4,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-24DOI: 10.1016/j.sigpro.2025.110058
Wenjuan Shi, Xiangwei Zheng, Lifeng Zhang, Cun Ji, Yuang Zhang, Ji Bian
{"title":"Multi-Object Tracking based on Optimal Transport and Coordinate Attention Mechanism","authors":"Wenjuan Shi, Xiangwei Zheng, Lifeng Zhang, Cun Ji, Yuang Zhang, Ji Bian","doi":"10.1016/j.sigpro.2025.110058","DOIUrl":"10.1016/j.sigpro.2025.110058","url":null,"abstract":"<div><div>Multi-Object Tracking (MOT) has currently attracted significant interest due to its wide applications in various fields, such as autonomous driving, intelligent surveillance, and behavior recognition. However, appearance similarity of different objects results in low accuracy of target matching and difficulties in data association. In this paper, we propose a Multi-Object Tracking based on Optimal Transport and Coordinate Attention Mechanism (MOT2A), which addresses above challenges by integrating the attention mechanism with optimal transport. These strategies effectively enhance the extraction of discriminative appearance features and improve target matching between different frames. Firstly, we construct a novel Coordinate attention module (CASA), which models the interdependence between the channel domain and the spatial domain of the feature map. Secondly, a Triplet loss with optimal transport (SK-Triplet) is designed to adjust the distance matrix for effective clustering of positive and negative samples during loss calculation. Finally, extensive experiments are conducted on MOT17 and MOT20. For MOT17: 79.4 MOTA, 78.9 IDF1, and 63.9 HOTA; For MOT20: 77.0 MOTA, 76.3 IDF1, and 62.3 HOTA are achieved, respectively. Compared to existing MOT methods, our method shows significant improvements in accuracy and stability. The code is available at: <span><span>https://github.com/420-s/MOT2A</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110058"},"PeriodicalIF":3.4,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-21DOI: 10.1016/j.sigpro.2025.110061
Yang Song, Jincan Zhang, Jinli Chen, Gangyi Tu, Jiaqiang Li
{"title":"Outlier-resistant Bayesian tensor completion for angle estimation in bistatic MIMO radar under array element failures","authors":"Yang Song, Jincan Zhang, Jinli Chen, Gangyi Tu, Jiaqiang Li","doi":"10.1016/j.sigpro.2025.110061","DOIUrl":"10.1016/j.sigpro.2025.110061","url":null,"abstract":"<div><div>Conventional angle estimation algorithms for multiple-input multiple-output (MIMO) radar are susceptible to array element failures and impulsive noise, which makes achieving accurate estimates in practical applications challenging. To remedy this, we propose an outlier-resistant Bayesian tensor completion algorithm for joint direction-of-departure (DOD) and direction-of-arrival (DOA) estimation in bistatic MIMO radar under element failures and impulsive noise. First, we constructed a slice-missing tensor signal model that is corrupted by outliers. To achieve better low-rank regularization on this tensor, we convert it into a structured tensor with randomly missing entries. We then design an outlier-resistant Bayesian tensor completion model, which accounts for array element failures and the \"heavy-tailed\" nature of impulsive noise. In the proposed model, the reconstruction of missing entries represents array element failures, while Student-t distribution models the impulsive noise in the measurements. A variational Bayesian inference scheme is developed to address the proposed model, which alternates among estimating the factor matrices, recovering the tensor rank, and mitigating impulsive noise. Finally, the completed factor matrix is used to extract DODs and DOAs using the shift invariance technique. Simulation results confirm the outstanding performance of the proposed algorithm in estimating target numbers and angles under element failures and impulsive noise.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110061"},"PeriodicalIF":3.4,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-19DOI: 10.1016/j.sigpro.2025.110068
Yongjian Li , Meng Chen , Xinliang Qu , Baokun Han , Lei Liu , Shoushui Wei
{"title":"An atrial fibrillation signals analysis algorithm in line with clinical diagnostic criteria","authors":"Yongjian Li , Meng Chen , Xinliang Qu , Baokun Han , Lei Liu , Shoushui Wei","doi":"10.1016/j.sigpro.2025.110068","DOIUrl":"10.1016/j.sigpro.2025.110068","url":null,"abstract":"<div><div>The detection of atrial fibrillation using deep learning techniques is a hot topic in the field of signal processing. However, simply stacking modules to pursue accuracy, or compressing inputs and parameters to pursue real-time performance, leads to gambling problem between information redundancy and information loss in deep learning algorithms. At the same time, the features obtained by deep learning lack interpretability. Therefore, this study proposes a T neural network (T-Net) that integrates feature extraction, selection, and fusion. In T-Net, horizontal path extracts multi-scale information of electrocardiograms through multi-level feature reuse, feature filter embeds attention mechanism and voting algorithm internally to select information flow, and vertical path uses channel-wise point-to-point weighting to capture the nonlinear relationships of multi-scale information. Through pre-training and fine-tuning on the MIT-BIH atrial fibrillation database consisting of 23 patients, and testing on the Shandong Provincial Hospital database consisting of 252 patients, T-Net achieved accuracy, specificity, sensitivity, and F1 score of 97.95 %, 97.01 %, 98.89 %, and 97.97 %, respectively. T-Net addresses the gambling problem between information redundancy and information insufficiency, and the extracted features demonstrate good interpretability consistent with clinical diagnostic criteria, showing promising clinical application prospects.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110068"},"PeriodicalIF":3.4,"publicationDate":"2025-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-17DOI: 10.1016/j.sigpro.2025.110044
Mingjing Cui , Yunxiang Jiang , Dongyuan Lin , Yunfei Zheng , Shiyuan Wang
{"title":"Robust mixture filtering block based on logarithmic Student’s t-based criterion","authors":"Mingjing Cui , Yunxiang Jiang , Dongyuan Lin , Yunfei Zheng , Shiyuan Wang","doi":"10.1016/j.sigpro.2025.110044","DOIUrl":"10.1016/j.sigpro.2025.110044","url":null,"abstract":"<div><div>Determining an appropriate cost function is crucial to develop adaptive filters. However, current robust algorithms may not be capable of satisfying the requirements of various non-Gaussian environments due to their limited performance surfaces and gradient relationships. To this end, a novel and robust cost function called logarithmic Student’s <span><math><mi>t</mi></math></span>-based (logST) criterion using Student’s <span><math><mi>t</mi></math></span> distribution is first proposed in this paper. Owing to its excellent properties in the algorithm generalization, robust gradient relationship, and efficient performance surface, the proposed logST algorithm achieves filtering accuracy improvement in both Gaussian and non-Gaussian environments. To further enhance the convergence performance and tracking capability in nonlinear system identification, a novel nonlinear block-oriented framework is constructed using the mixture of original space and random Fourier features space. Then, a recursive method is applied to achieve optimization solution in this nonlinear block-oriented framework, generating the mixture random Fourier features recursive logST (MRFFRlogST) algorithm. Finally, simulations on linear and nonlinear system identifications, as well as Chua’s time-series prediction under various noise environments validate the superiorities of logST and MRFFRlogST.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110044"},"PeriodicalIF":3.4,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-17DOI: 10.1016/j.sigpro.2025.110057
Enping Lin , Ze Fang , Yuqing Huang , Yu Yang , Zhong Chen
{"title":"Non-uniform sampling reconstruction for symmetrical NMR spectroscopy by exploiting inherent symmetry","authors":"Enping Lin , Ze Fang , Yuqing Huang , Yu Yang , Zhong Chen","doi":"10.1016/j.sigpro.2025.110057","DOIUrl":"10.1016/j.sigpro.2025.110057","url":null,"abstract":"<div><div>Symmetrical NMR spectroscopy constitutes a vital branch of multidimensional NMR spectroscopy, providing a powerful tool for the structural elucidation of biological macromolecules. Non-Uniform Sampling (NUS) serves as an effective strategy for averting the prohibitive acquisition time of multidimensional NMR spectroscopy by only sampling a few points according to NUS sampling schedules and reconstructing missing points via algorithms. However, current sampling schedules are unable to maintain the accurate recovery of cross peaks that are weak but important. In this work, we propose a novel sampling schedule— SCPG (Symmetrical Copy Poisson Gap) and employ CS (Compressed Sensing) methods for reconstruction. We theoretically prove that the symmetrical constraint, apart from sparsity, is implicitly implemented when SCPG is combined with CS methods. The simulated and experimental data substantiate the advantage of SCPG over state-of-the-art 2D Woven PG in the NUS reconstruction of symmetrical NMR spectroscopy.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110057"},"PeriodicalIF":3.4,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-17DOI: 10.1016/j.sigpro.2025.110042
Xiangqing Xiao , Hua Wang , Jinfeng Hu , Xin Tai , Yongfeng Zuo , Huiyong Li , Kai Zhong , Dongxu An
{"title":"Unimodular waveform design for ambiguity function shaping with spectral constraint via a manifold-based exact penalty method","authors":"Xiangqing Xiao , Hua Wang , Jinfeng Hu , Xin Tai , Yongfeng Zuo , Huiyong Li , Kai Zhong , Dongxu An","doi":"10.1016/j.sigpro.2025.110042","DOIUrl":"10.1016/j.sigpro.2025.110042","url":null,"abstract":"<div><div>Unimodular waveform design for ambiguity function (AF) shaping is a key technology in radar system. Depending on whether the spectral compatibility of the waveform is considered or not, the problem can be categorized into two forms, the first without imposing spectral constraints and the second with spectral constraints. Both of them are complex quartic non-convex problems with constant modulus constraint (CMC), which are challenging to be solved. It is worth noting that the second one obtains more attention due to the improved spectral compatibility, but the involved inequality constraints further increase the difficulty of solving this problem. We note that the inequality constraints can be transformed into penalty terms of the objective function by constructing an exact penalty function. Additionally, the complex circle manifold (CCM) naturally satisfies the CMC, providing a suitable framework to address the resultant problem. Based on the above considerations, we propose a manifold-based exact penalty method (MEP). First, the inequality constraints are eliminated by constructing them as exact penalty terms of the objective function using an exact penalty function, which incorporates the constraints into the objective function. The resulting problem is then projected onto the CCM, transforming the problem into a complex quartic unconstrained optimization problem. Finally, a conjugate gradient (CG) method is derived to solve this unconstrained problem on the CCM. Simulation results confirm that compared to the methods in Wu et al. (2017), Alhujaili et al. (2020) and Zhang et al. (2023), the proposed method exhibits the following superior performance:(1) higher signal-to-interference ratios (SIR), (2) deeper notches and more precise control of the stopbands level of energy spectrum distribution (ESD), (3) stronger detection performance to weak targets.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110042"},"PeriodicalIF":3.4,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143848281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-16DOI: 10.1016/j.sigpro.2025.110034
Jingang Wang, Songbin Li, Ke Shi
{"title":"Radar target tracking based on motion characteristic and distribution pattern matching","authors":"Jingang Wang, Songbin Li, Ke Shi","doi":"10.1016/j.sigpro.2025.110034","DOIUrl":"10.1016/j.sigpro.2025.110034","url":null,"abstract":"<div><div>The mitigation of false alarm rates under real-world radar operating conditions represents a critical challenge in advancing radar target detection algorithms. This study proposes that utilizing multi-frame correlation information through radar target tracking constitutes an effective solution for suppressing false alarms. We present a radar target tracking methodology that integrates motion characteristic and distribution pattern matching, effectively leveraging multi-frame radar measurements and echo amplitude information. This approach enables false alarm reduction through trajectory consistency validation. Specifically, the method operates in two stages: First, state filtering based on motion characteristics is applied to predict potential candidate regions for previous detections within the current frame. Subsequently, within these candidate regions, a deep learning-based similarity evaluation framework employing a self-supervised Siamese network performs distribution pattern matching to establish optimal data associations. Experimental validation demonstrates that the proposed method achieves a 5.82% improvement in F1-score over benchmark algorithms, confirming its enhanced detection reliability and operational effectiveness.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110034"},"PeriodicalIF":3.4,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143851599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-16DOI: 10.1016/j.sigpro.2025.110045
Jing Zhou, Zhichao Zhang
{"title":"Dynamic and static Lagrange multiplier μ based Bayesian CP factorization with orthogonal factors: Theory and applications","authors":"Jing Zhou, Zhichao Zhang","doi":"10.1016/j.sigpro.2025.110045","DOIUrl":"10.1016/j.sigpro.2025.110045","url":null,"abstract":"<div><div>The existing <span><math><mi>μ</mi></math></span>-Singular Value Decomposition (<span><math><mi>μ</mi></math></span>-SVD) denoising algorithm is capable of extracting gear fault information under strong noise conditions. However, this algorithm is only applicable to two-dimensional real-valued data and lacks a mechanism for implementing Automatic Rank Determination (ARD) in high-dimensional data. In this paper, a Bayesian and Tensor treatment of <span><math><mi>μ</mi></math></span>-SVD is employed to enable ARD. To further investigate the impact of the Lagrange multiplier <span><math><mi>μ</mi></math></span> on the proposed <span><math><mi>μ</mi></math></span>-variational Bayesian (<span><math><mi>μ</mi></math></span>-VB) algorithm, we examine its performance from both static and dynamic perspectives. Simulation results demonstrate that the <span><math><mi>μ</mi></math></span>-VB algorithm achieves ARD and performs well in noise reduction. Further the <span><math><mi>μ</mi></math></span>-VB algorithm performs better in wireless communication and linear image coding across numerical domains, tensor sizes, orthogonal factors, and <span><math><mi>μ</mi></math></span> settings.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110045"},"PeriodicalIF":3.4,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143848301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Signal ProcessingPub Date : 2025-04-15DOI: 10.1016/j.sigpro.2025.110056
Yijie Lin , Jui‐Chuan Liu , Ching-Chun Chang , Chin‐Chen Chang
{"title":"A puzzle matrix oriented secret sharing scheme for dual images with reversibility","authors":"Yijie Lin , Jui‐Chuan Liu , Ching-Chun Chang , Chin‐Chen Chang","doi":"10.1016/j.sigpro.2025.110056","DOIUrl":"10.1016/j.sigpro.2025.110056","url":null,"abstract":"<div><div>Dual image reversible data hiding is a specialized research direction in the field of information security. Secret messages are embedded into the original image using an image signal processing algorithm for reversible data hiding. This process generates two share images, indistinguishable from the original image, and each distributed to different receivers. Only through collaboration between the receivers can the original image and secret message be fully recovered. A dual image reversible data hiding scheme is proposed based on a reference matrix. The pixel pairs from the two share images collaborate and result in embedding 6 bits using the puzzle matrix; thus, achieving an embedding rate of up to 1.5 bits per pixel. Experimental results demonstrate that the proposed scheme offers a high embedding capacity of up to 786,432 bits, good visual quality, and fast execution time. Compared to other state-of-the-art dual image reversible data hiding schemes, our advantage mainly lies in the larger embedding capacity. Compared with a previously proposed scheme with the same embedding capacity, our proposed scheme provides better visual quality and faster execution time.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"236 ","pages":"Article 110056"},"PeriodicalIF":3.4,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143851598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}