检索结果-内蒙古大学图书馆

Modified PUMA/EPUMA Based on Forward and Backward Linear Prediction for DOA Estimation

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2025年 22卷

作者： Ma, Biyun Zhu, Fu Wang, Yide Zhu, Qingqing Liu, Jiaojiao South China Univ Technol Sch Elect & Informat Engn Guangzhou 510640 Peoples R China Minist Nat Resources Key Lab Marine Environm Survey Technol & Applicat Guangzhou 510300 Peoples R China Nantes Univ Inst Elect & Technol NumeR IETR CNRS UMR 6164 F-44306 Nantes France

The principal-singular-vector utilization for modal analysis (PUMA) and its modification (Mod-PUMA), which utilize forward linear prediction (FLP) to process the signal subspace, experience significant performance degradation if there are multiple coherent sources and such a performance degradation will be further aggravated in low-SNR regions, which is primarily attributed to the outliers arising from inaccurate estimations of the signal subspace. To address these issues, we propose an extension version of PUMA-related algorithms, called FBLP-Mod-PUMA/enhanced-PUMA (EPUMA). The proposed algorithms improve the threshold performance by refining the signal subspace through forward and backward linear prediction (FBLP), effectively mitigating subspace leakage when dealing with coherent sources. The number of resolvable coherent sources has been theoretically derived and simulation results are provided to show the performance of the proposed algorithms.

关键词： Covariance matrices Eigenvalues and eigenfunctions Vectors Prediction algorithms Direction-of-arrival estimation Smoothing methods Simulation signal to noise ratio signal processing algorithms Sensor arrays Coherent sources direction-of-arrival (DOA) estimation forward and backward linear prediction (FBLP) principal-singular-vector utilization for modal analysis (PUMA)/enhanced-PUMA (EPUMA)

来源：评论

学校读者我要写书评

暂无评论

Consensus Iterated Posterior Linearization Filter for Distributed State Estimation

引用

IEEE signal processing LETTERS 2025年 32卷 561-565页

作者： Garcia-Fernandez, Angel F. Battistelli, Giorgio Univ Liverpool Dept Elect Engn & Elect Liverpool L69 3GJ England Univ Firenze Dipartimento Ingn Informaz DINFO I-50139 Florence Italy

This paper presents the consensus iterated posterior linearisation filter (IPLF) for distributed state estimation. The consensus IPLF algorithm is based on a measurement model described by its conditional mean and covariance given the state, and performs iterated statistical linear regressions of the measurements with respect to the current approximation of the posterior to improve estimation performance. Three variants of the algorithm are presented based on the type of consensus that is used: consensus on information, consensus on measurements, and hybrid consensus on measurements and information. Simulation results show the benefits of the proposed algorithm in distributed state estimation.

关键词： Covariance matrices Approximation algorithms Vectors Kalman filters signal processing algorithms Sensors Consensus algorithm State estimation Current measurement Noise measurement Consensus distributed state estimation iterated posterior linearisation nonlinear filtering

来源：评论

学校读者我要写书评

暂无评论

Multiple Dimensional Correntropy Kalman Filter

引用

IEEE signal processing LETTERS 2025年 32卷 1011-1015页

作者： Wang, Guoqing Zhu, Zhaolei Yang, Chunyu Ma, Lei China Univ Min & Technol Sch Informat & Control Engn Xuzhou 221116 Peoples R China

The letter addresses the robust state estimation problem of non-Gaussian systems disturbed by outliers. Unlike the existing correntropy-based state estimation framework, which uses a uniform weight for the evaluated error vectors and solely relies on inaccurate nominal covariance matrices for estimating the system state, this work proposes a novel maximum-correntropy Kalman filter. This new approach utilizes multiple dimensional correntropy to assess the similarity between vectors across different dimensions. Additionally, it adjusts the covariance matrices simultaneously by utilizing the adopted matrix similarity measure within the modified correntropy framework. Simulations on target tracking demonstrate that our proposed algorithm exhibits excellent estimation accuracy and robustness while possessing adaptive capability even under time-varying heavy-tailed noises.

关键词： Vectors Noise Covariance matrices signal processing algorithms Kernel Pollution measurement Target tracking Noise measurement Kalman filters Accuracy Multiple dimension correntropy non-Gaussian noise robust Kalman filter covariance estimation

来源：评论

学校读者我要写书评

暂无评论

Minimum Total Quaternion Error Entropy Filtering With Fiducial Points Against Asymmetric Noise

引用

IEEE signal processing LETTERS 2025年 32卷 306-310页

作者： Lin, Dongyuan Cai, Peng Chen, Xiaofeng Zheng, Yunfei Wang, Shiyuan Southwest Univ Coll Elect & Informat Engn Chongqing 400715 Peoples R China Chongqing Jiaotong Univ Dept Math Chongqing 400074 Peoples R China

Quaternion adaptive filters (QAFs) are extensively used in processing three- or four-dimensional signals effectively. However, their performance can significantly deteriorate or even diverge when system inputs and outputs are contaminated by complex noises. Therefore, this letter addresses the issue of parameter estimation in the quaternion errors-in-variables (QEIV) in asymmetric noise. First, a novel robust criterion, called improved quaternion minimum error entropy criterion with fiducial points (IQMEEF), is constructed. Then, a minimum total quaternion error entropy algorithm with fiducial points (MTQEEF) is proposed by integrating the IQMEEF criterion with the total least squares (TLS) method, leveraging stochastic gradient and quaternion generalized Hamilton-real (GHR) calculus theory. Finally, simulations validate the superior performance of MTQEEF in the QEIV model under asymmetric noise environments.

关键词： Quaternions Noise Entropy signal processing algorithms Adaptation models Vectors Parameter estimation Kernel Filtering algorithms Calculus Robust quaternion adaptive filters quaternion error entropy criterion with fiducial points quaternion errors-in-variables model asymmetric noise

来源：评论

学校读者我要写书评

暂无评论

Sparse Projection Matrix Approximation and Its Applications

引用

IEEE signal processing LETTERS 2025年 32卷 351-355页

作者： Zhai, Zheng Wu, Mingxin Xu, Jialu Li, Xiaohui Beijing Normal Univ Fac Arts & Sci Dept Stat Zhuhai 519088 Peoples R China Yantai Univ Sch Math & Informat Sci Yantai 264032 Peoples R China

This letter introduces a sparse regularized projection matrix approximation (SPMA) model to recover cluster structures from affinity matrices. The model is formulated as a projection approximation problem with an entry-wise sparsity penalty to encourage sparse solutions. We propose two algorithms to solve this problem: one involves direct optimization on the Stiefel manifold using the Cayley transformation, while the other employs the Alternating Direction Method of Multipliers (ADMM). Numerical experiments on synthetic and real-world datasets demonstrate that our regularized projection matrix approximation approach significantly outperforms state-of-the-art methods in clustering accuracy and performance.

关键词： Sparse matrices Manifolds Approximation algorithms signal processing algorithms Optimization Convex functions Matrix decomposition Human activity recognition Eigenvalues and eigenfunctions Clustering algorithms ADMM clustering regularisation Stiefel manifold sparse projection matrix approximation

来源：评论

学校读者我要写书评

暂无评论

A Hardware-Friendly Lightweight Partition Decision Algorithm for VVC Intra and Inter Coding

引用

IEEE signal processing LETTERS 2025年 32卷 941-945页

作者： Zan, Zhao Huang, Leilei Chen, Shushi Zeng, Xiaoyang Fan, Yibo Fudan Univ State Key Lab Integrated Chips & Syst Shanghai 201203 Peoples R China East China Normal Univ Inst Microelect Circuits & Syst Shanghai 200241 Peoples R China

The Versatile Video Coding (VVC) standard notably enhances encoding efficiency with the Quad-Tree plus Multi-Type Tree (QTMTT) partition structure. However, the complex QTMTT tool presents substantial challenges in both software and hardware implementation. To overcome those challenges, this paper introduces a hardware-friendly partition decision algorithm for VVC intra and inter coding. Firstly, we propose a lightweight backbone network to extract partition-aware features. Secondly, we employ a Quantisation Parameter (QP) fusion network to regulate the impact of QPs on the partition structure. Additionally, we apply a top-down threshold-driven post-processing algorithm, in which improbable partition types are removed to directly derive the unique partition structure. Experiments show that our method not only exceeds the previous state-of-the-art work in BD-BR performance, but also shows sufficient hardware-friendly characteristics. To the best of our knowledge, this work is among the earliest to comprehensively discuss and implement a hardware-friendly partition decision algorithm.

关键词： Partitioning algorithms Hardware Encoding signal processing algorithms Prediction algorithms Feature extraction Complexity theory Software algorithms Convolution Software VVC QTMTT partition decision complexity reduction hardware-friendly algorithm convolutional neural network intra coding inter coding

来源：评论

学校读者我要写书评

暂无评论

Bias-Compensated Normalized Iterative Wiener Filter Algorithm With Noisy Input

引用

IEEE signal processing LETTERS 2025年 32卷 1445-1449页

作者： Yuan, Hai Lu, Lu Zhu, Guangya Chen, Badong Sichuan Univ Coll Elect & Informat Engn Chengdu 610065 Peoples R China Sichuan Univ Coll Elect Engn Chengdu 610065 Peoples R China Xi An Jiao Tong Univ Inst Artificial Intelligence & Robot Xian 710049 Peoples R China

The iterative Wiener filter (IWF) algorithm can achieve a fast convergence rate. However, its performance may degrade when it encounters noisy input scenarios. To tackle this problem, a novel IWF algorithm incorporating bias-compensation (BC-IWF) is proposed, which can enhance the performance of the algorithm by estimating the input noise variance. The BC-IWF algorithm optimizes the step size for each iteration and updates along the direction of the gradient. To further reduce the steady-state error, a normalized IWF by making use of the bias-compensation scheme (BC-NIWF) algorithm is proposed. Moreover, the steady-state performance of the BC-NIWF algorithm is analyzed. Simulation results demonstrate the validity of the theoretical analysis and the BC-NIWF algorithm achieves improved misadjustment compared with the state-of-the-art algorithms.

关键词： signal processing algorithms Approximation algorithms Noise measurement Steady-state Vectors Wiener filters Convergence Artificial intelligence Training Simulation Bias compensation iterative Wiener filter noisy input model steady-state analysis

来源：评论

学校读者我要写书评

暂无评论

FM-RTDETR: Small Object Detection Algorithm Based on Enhanced Feature Fusion With Mamba

引用

IEEE signal processing LETTERS 2025年 32卷 1570-1574页

作者： Yang, Yuchuan Dai, Jiahui Wang, Yong Chen, Yafei Chongqing Univ Technol Sch Artificial Intelligence Chongqing 401135 Peoples R China

Traditional real-time object detection networks deployed in autonomous aerial vehicles (AAVs) struggle to extract features from small objects in complex backgrounds with occlusions and overlapping objects. To address this challenge, we propose FM-RTDETR, a real-time object detection algorithm optimized for small object detection. We redesign the encoder of RT-DETRv2 by integrating the Feature Aggregation and Diffusion Network (FADN), improving the algorithm's ability to capture contextual information. Subsequently, we introduce the Parallel Atrous Mamba Feature Fusion Module (PAMFFM), which combines shallow and deep semantic information to better capture small object features. Furthermore, we propose the Cross-stage Enhanced Feature Fusion Module (CEFFM), merging features for small objects to provide richer and more detailed information. Finally, we propose STIoU Loss, which incorporates a penalty term to adjust the scaling of the loss function, improving detection granularity for small objects. FM-RTDETR achieves AP(50) scores of 54.0% and 56.3% on the VisDrone2019-DET and AI-TOD datasets. Compared with other state-of-the-art methods, our method shows great potential in small object detection from drones.

关键词： Object detection signal processing algorithms Feature extraction Transformers Training Real-time systems Accuracy Decoding Convolution Artificial intelligence Feature fusion RT-DETRv2 small object detection VisDrone

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Sensor Scheduling for State Estimation Over Homogeneous Multi-Hop Networks

引用

IEEE signal processing LETTERS 2025年 32卷 411-415页

作者： Li, Yao Song, Rui Zheng, Jianyong Guan, Xinping Shanghai Univ Sch Future Technol Shanghai 200444 Peoples R China Shanghai Jiao Tong Univ Dept Automat Shanghai 200240 Peoples R China

In this letter, anestimation-oriented power-constrained sensor scheduling problem over multi-hop sensor networks is studied. Two different online scheduling schemes for multi-hop transmission, i.e., global-delay-based scheduling (GS) and covariance-based scheduling (CS) are proposed, respectively. We propose a stochastic triggering scheme to satisfy power constraints. A Markovian model is adopted to formulate the state transition relationship in the scheduling process. In order to calculate the switching threshold and selection probabilities, an algorithm for parameters determination is further properly designed to overcome the coupling property between hops. We have explicitly analyzed the performance of GS by figuring out the upper and lower bounds of cost. Moreover, the superiority and optimality of CS have been theoretically proved by using the optimal state distribution method. Numerical simulations and comparisons with existing methods have been conducted illustratively to verify the correctness and effectiveness of our proposed schemes, algorithms and results.

关键词： Spread spectrum communication Schedules Prediction algorithms Switches Processor scheduling Wireless sensor networks Lower bound signal processing algorithms Relays Costs Multi-hop networks sensor scheduling state estimation

来源：评论

学校读者我要写书评

暂无评论

Hypformer: A Fast Hypothesis-Driven Rescoring Speech Recognition Framework

引用

IEEE signal processing LETTERS 2025年 32卷 471-475页

作者： Zhuang, Xuyi Qian, Yukun Wang, Mingjiang Harbin Inst Technol Shenzhen Shenzhen 518000 Peoples R China

Recently, the performance of non-autoregressive ASR models has made significant progress but still lags behind hybrid CTC/attention systems. This paper introduces Hypformer, a fast hypothesis-driven rescoring speech recognition framework. Multiple hypothetical prefixes are realized by fast prefix generation algorithm. With two different rescoring methods, nar-ar rescoring and nar$<^>{2}$ rescoring, Hypformer can flexibly switch between autoregressive and non-autoregressive decoding modes to perform rescoring of hypothesis prefixes. Experiments on the standard Mandarin datasets AISHELL-1 and AISHELL-2 demonstrate that Hypformer outperforms the state-of-the-art Hybrid CTC/Attention systems in ASR performance while achieving a speedup of over six times. Experiments on the Mandarin sub-dialect dataset KeSpeech indicate that Hypformer achieves more accurate recognition by leveraging richer contextual information.

关键词： Decoding Training Iterative decoding Transformers Accuracy Standards Symbols Switches signal processing algorithms Probability distribution ASR E2E non-autoregressive hypothesis prefixes rescoring

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：