Signal Processing最新文献

A clustering variational Bayesian Kalman filter with heavy-tailed measurement noise

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-04-01 DOI: 10.1016/j.sigpro.2025.110010

Gang Wang, Zuxuan Zhang, Haihao Yang, Zhoubin Yao

引用次数: 0

Cauchy–Gaussian maximum mixture correntropy Kalman filter with component-by-component construction

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-31 DOI: 10.1016/j.sigpro.2025.110008

Shungang Peng, Peng Cai, Dongyuan Lin, Yunfei Zheng, Shiyuan Wang

{"title":"Cauchy–Gaussian maximum mixture correntropy Kalman filter with component-by-component construction","authors":"Shungang Peng, Peng Cai, Dongyuan Lin, Yunfei Zheng, Shiyuan Wang","doi":"10.1016/j.sigpro.2025.110008","DOIUrl":"10.1016/j.sigpro.2025.110008","url":null,"abstract":"<div><div>The Kalman filter (KF) systematically optimizes the quasi-Monte Carlo (QMC) sampling points by employing a component-by-component (CBC) construction principle tailored to specific integration dimensions and accuracy requirements. This optimization enhances the approximation accuracy of Gaussian-weighted multidimensional integrals (GMIs) within the context of the KF. However, the KF based on CBC sampling points uses the minimum mean square error (MMSE) criterion, which can lead to significant estimation bias in the presence of non-Gaussian noises. To address this issue, this paper first proposes a novel Cauchy–Gaussian maximum mixture correntropy Kalman filter with component-by-component construction (CGMCKF-CBC) by the designed novel Cauchy–Gaussian maximum mixture correntropy (CGMC). Unlike the traditional maximum mixture correntropy criterion (MMCC), the CGMC uses the mixture of Cauchy kernel and Gaussian kernel as the cost function, and updates the posteriori estimates by the form of fixed-point iteration. Next, to further address the parameters selection issue existing in CGMCKF-CBC, an adaptive optimization strategy is proposed to determine appropriate parameters, resulting a variable CGMCKF-CBC (VCGMCKF-CBC). Then, the convergence analysis and complexity evaluation of CGMCKF-CBC are also conducted. Finally, two simulation examples are conducted in non-Gaussian noises environments to validate the excellent filtering accuracy and robustness of CGMCKF-CBC and VCGMCKF-CBC.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110008"},"PeriodicalIF":3.4,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143767814","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}

引用次数: 0

Optimal estimation of the canonical polyadic decomposition from low-rank tensor trains

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-28 DOI: 10.1016/j.sigpro.2025.110001

Clémence Prévost , Pierre Chainais

{"title":"Optimal estimation of the canonical polyadic decomposition from low-rank tensor trains","authors":"Clémence Prévost , Pierre Chainais","doi":"10.1016/j.sigpro.2025.110001","DOIUrl":"10.1016/j.sigpro.2025.110001","url":null,"abstract":"<div><div>Tensor factorization has been steadily used to represent high-dimensional data. In particular, the canonical polyadic decomposition (CPD) is very appreciated for its remarkable uniqueness properties. However, computing the high-order CPD is challenging: numerical issues and high needs for storage and processing can make algorithms diverge. Furthermore, the recovery of the CP factors is an ill-posed problem. One way to circumvent this limitation is to exploit the equivalence between the CPD and the Tensor Train Decomposition (TTD). This paper formulates the CPD as a dimension reduction using a TTD followed by a global marginally convex optimization problem. This global optimization scheme estimates the CP factors with minimal error. The resulting approach, Dimensionality Reduction, joint Estimation of the Ambiguity Matrices and the CP FACtors (DREAMFAC), relies on a block-coordinate descent that reaches a first-order stationary point when estimating the CP factors. DREAMFAC is also shown to be an optimal estimator that reaches the corresponding constrained Cramér–Rao bound. It therefore appears as a state-of-the-art solution to estimate the best rank-<span><math><mi>K</mi></math></span> CPD of a tensor (when it exists). Its performance is illustrated on the problem of parameter estimation in a dual-polarized MIMO system. Numerical experiments show the excellent practical performance of DREAMFAC, even with very low SNR.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110001"},"PeriodicalIF":3.4,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143737924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reversible adversarial visible image watermarking

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-26 DOI: 10.1016/j.sigpro.2025.109999

Xue Xie, Jun Jiang, Jiansong Zhang, Kejiang Chen, Weiming Zhang, Nenghai Yu

{"title":"Reversible adversarial visible image watermarking","authors":"Xue Xie, Jun Jiang, Jiansong Zhang, Kejiang Chen, Weiming Zhang, Nenghai Yu","doi":"10.1016/j.sigpro.2025.109999","DOIUrl":"10.1016/j.sigpro.2025.109999","url":null,"abstract":"<div><div>Visible watermarking serves as a crucial security mechanism for safeguarding the copyright of digital images. Recent advancements, however, have shown that deep neural networks can effectively remove these watermarks without altering the underlying host image, posing a substantial risk to copyright protection. Motivated by the susceptibility of neural networks to adversarial perturbations, various adversarial visible watermarking techniques have been introduced. Nonetheless, these approaches often overlook the need for image reversibility, which is vital for authorized sharing while maintaining privacy. To address this issue, we propose <strong>R</strong>eversible <strong>A</strong>dversarial <strong>V</strong>isible <strong>W</strong>atermarking (RAVW), which uses Gradient-weighted Class Activation Mapping (Grad-CAM) to pinpoint the important regions in the host image that are optimal for watermark embedding. It then employs an end-to-end generative model to create reversible adversarial visible watermarks within these regions, effectively counteracting watermark removal networks. Additionally, authorized users can eliminate the visible watermark via a dedicated restoration module. Comprehensive experimental evaluations confirm the robustness of our method in preserving visible watermarks and its effectiveness against watermark removal networks.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 109999"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143747455","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}

引用次数: 0

Residual-optimized general linear chirplet transform: A method for time–frequency feature extraction

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-26 DOI: 10.1016/j.sigpro.2025.110006

Fuzheng Liu , Chenglong Ye , Chang Peng , Xiangyi Geng , Mingshun Jiang , Lei Zhang , Faye Zhang

{"title":"Residual-optimized general linear chirplet transform: A method for time–frequency feature extraction","authors":"Fuzheng Liu , Chenglong Ye , Chang Peng , Xiangyi Geng , Mingshun Jiang , Lei Zhang , Faye Zhang","doi":"10.1016/j.sigpro.2025.110006","DOIUrl":"10.1016/j.sigpro.2025.110006","url":null,"abstract":"<div><div>Time–frequency analysis (TFA) is an essential sub-area in signal processing, and many of its methods are widely used in various fields. Current methods cannot address the multi-component signals well when facing strong noise, which makes it difficult to describe the accurate time–frequency trajectory. This paper proposes a TFA method named residual optimized general linear chirplet transform (ROGLCT) to achieve better time–frequency energy concentration and robustness. Firstly, the beetle antennae search (BAS) algorithm with Rényi entropy was utilized as the fitness function to optimize GLCT parameters. Then, the maximum energy modes of the optimized GLCT time–frequency distribution are iteratively separated. Finally, the separated modes are squeezed to rebuild the spectrum. Numerical simulation and actual signals (echolocation and vibration signals) verify the proposed method’s effectiveness. Compared with other advanced methods, ROGLCT can overcome noise interference and depict clear, energy-concentrated time–frequency representation (TFR) in complex environments.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110006"},"PeriodicalIF":3.4,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143747456","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}

引用次数: 0

Constrained coupled CPD of complex-valued multi-slice multi-subject fMRI data

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-25 DOI: 10.1016/j.sigpro.2025.110004

Li-Dan Kuang, Hao Zhu, Lei Long, Ting Tang, Yan Gui, Jin Zhang

{"title":"Constrained coupled CPD of complex-valued multi-slice multi-subject fMRI data","authors":"Li-Dan Kuang, Hao Zhu, Lei Long, Ting Tang, Yan Gui, Jin Zhang","doi":"10.1016/j.sigpro.2025.110004","DOIUrl":"10.1016/j.sigpro.2025.110004","url":null,"abstract":"<div><div>Considering that the four-way complex-valued multi-subject fMRI tensor inevitably contains the large unwanted brain-out voxels, this paper innovatively combines multi-subject brain-in fMRI data with same slices as three-way multi-slice multi-subject fMRI tensors and thus brain-out voxels are discarded. Additionally, adjacent-slice fMRI tensors can be further merged, and multi-slice multi-subject fMRI tensors of <em>N</em> groups are formed, and are jointly decomposed by a novel spatiotemporally constrained coupled canonical polyadic decomposition (CCPD) by sharing temporal and subject modes but allowing slice-group differences. The spatial phase sparsity and orthonormality constraints are added on rank-<em>R</em> least-squares fit of <em>N</em>-slice-group shared spatial maps (SMs) to reduce noise effect, cross-talk among components and inter-subject spatial variability which naturally occur in complex-valued fMRI data. To alleviate CCPD model and allow inter-subject temporal variability, the alternating shift-invariant rank-1 least-squares optimization is performed to update shared time courses (TCs), subject-specific time delays and intensities. Results of simulated and experimental fMRI analyses demonstrate that the proposed methods with different <em>N</em> groups outperformed the competing methods by 6.21 %∼23.61 % in terms of shared task-related SMs and TCs. The proposed methods with <em>N</em>=45 and <em>N</em>=3 respectively obtain the best performance in the presence of strong noise levels and slightly strong noise levels cases.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110004"},"PeriodicalIF":3.4,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143679408","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}

引用次数: 0

Total complex kernel risk-sensitive loss for robust DOA estimation

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-24 DOI: 10.1016/j.sigpro.2025.110002

Yijie Tang , Guobing Qian , Ke Wang , Hao Zeng

{"title":"Total complex kernel risk-sensitive loss for robust DOA estimation","authors":"Yijie Tang , Guobing Qian , Ke Wang , Hao Zeng","doi":"10.1016/j.sigpro.2025.110002","DOIUrl":"10.1016/j.sigpro.2025.110002","url":null,"abstract":"<div><div>Adaptive filtering-based approaches have been proposed for low-complexity direction of arrival (DOA) estimation. Nonetheless, existing methods demonstrate significant performance deterioration in bias compensation models that incorporate impulse noise. To address this challenge, we have introduced a novel similarity measure in kernel space, termed total complex kernel risk-sensitive loss (TCKRSL), which effectively extracts higher-order statistics from the data to mitigate the detrimental effects of outliers caused by impulse noise. Subsequently, we derived a robust adaptive filtering algorithm known as the minimum total complex kernel risk-sensitive loss (MTCKRSL) algorithm based on stochastic gradient descent and applied it to DOA estimation via adaptive nulling array antenna. To further enhance estimation performance, we implemented a variable step size (VSS) mechanism grounded in cumulative instantaneous error and the estimated signal power aimed at balancing the trade-off between steady-state error and convergence speed, resulting in the VSS-MTCKRSL algorithm. Additionally, the convergence properties and computational complexity of the proposed algorithm were analyzed elaborately. Simulation results across various performance metrics demonstrate that the proposed VSS-MTCKRSL algorithm outperforms the state-of-the-art algorithms regardless of the presence of impulse noise or Gaussian noise.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110002"},"PeriodicalIF":3.4,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716179","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}

引用次数: 0

Learning state and proposal dynamics in state-space models using differentiable particle filters and neural networks

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-21 DOI: 10.1016/j.sigpro.2025.109998

Benjamin Cox , Santiago Segarra , Víctor Elvira

引用次数: 0

Learning-based robust direction-of-arrival estimation with array imperfections

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-21 DOI: 10.1016/j.sigpro.2025.110000

Jiajing Chen , Rui Xiao , Yixin Jiang , Mingyi You , Qingjiang Shi , Xiang Cheng , Xuesong Cai

{"title":"Learning-based robust direction-of-arrival estimation with array imperfections","authors":"Jiajing Chen , Rui Xiao , Yixin Jiang , Mingyi You , Qingjiang Shi , Xiang Cheng , Xuesong Cai","doi":"10.1016/j.sigpro.2025.110000","DOIUrl":"10.1016/j.sigpro.2025.110000","url":null,"abstract":"<div><div>Estimating the direction-of-arrival (DOA) of signals is a fundamental problem in radar applications and source localization. With practical antenna arrays, the performance of the DOA estimation often degrades significantly in low-cost systems due to the presence of numerous imperfect factors. This paper presents novel deep learning-based approaches for DOA estimation in two-dimensional (2-D) scenarios with imperfect factors. As the selection between using in-phase and quadrature (I/Q) sequences or covariance vector methods is a well-known open issue in DOA estimation, we investigate the performance of deep learning-based networks using these two different inputs. Our study aims to provide insights into which approach is more effective in achieving high accuracy for DOA estimation under non-ideal array conditions. We first train a ResNet using I/Q sequences for DOA estimation and then propose a novel approach using a convolutional attention neural network (CANN) with a covariance vector as input, incorporating frequency information to enhance network robustness. Furthermore, we derive the Cramér-Rao Lower Bounds (CRLBs) of mean squared errors (MSEs) for parameter estimators in a single-source scenario. Simulations are conducted to evaluate the accuracy of the proposed DOA estimation approach, demonstrating its superior performance over existing techniques and its close approximation to the corresponding CRLBs.</div></div>","PeriodicalId":49523,"journal":{"name":"Signal Processing","volume":"234 ","pages":"Article 110000"},"PeriodicalIF":3.4,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143760835","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}

引用次数: 0

Multi-target elliptic positioning via difference of convex functions programming

IF 3.4 2区工程技术

Signal Processing Pub Date : 2025-03-20 DOI: 10.1016/j.sigpro.2025.109996

Xudong Dang, Hongwei Liu, Junkun Yan

引用次数: 0