检索结果-内蒙古大学图书馆

Low-Complexity Data-Driven Communication Neural Receivers

IEEE ACCESS 2025年 13卷 9325-9334页

作者： Wu, Qingle Liang, Yuanhui Ng, Benjamin K. Lam, Chan-Tong Ma, Yan Sichuan Univ Sci & Engn Sch Automat & Informat Engn Zigong 643000 Peoples R China Macao Polytech Univ Fac Appl Sci Macau 999078 Peoples R China Artificial Intelligence Key Lab Sichuan Prov Yibin 644000 Peoples R China Beijing Univ Posts & Telecommun BUPT Network Informat Ctr Beijing 100876 Peoples R China

Communication receivers based on deep learning have become a research hotspot in recent years. However, the excessive computational complexity and storage complexity prevent it from being deployed on communication hardware with limited resources. In order to reduce the computational complexity and required storage resources of communication neural receivers based on deep learning, we propose to use candecomp parafac (CP), Tucker, tensor-train (TT) and tensor-ring (TR) decomposition respectively to compress the data-driven deep learning based communication neural receiver. Through compression, the storage resources required by the communication neural receiver are reduced by about half with bit error rate (BER) performance degradation of only 0.75dB to 1.3dB.

关键词： Receivers Tensors Decoding OFDM Hardware Deep learning Symbols signal processing algorithms Computer architecture Wireless communication Tensor-train decomposition tensor-ring decomposition computational complexity communication neural receiver storage complexity

来源：评论

学校读者我要写书评

暂无评论

A Wideband Lateral Multibeam Tiling Synthetic Aperture Method for High-Speed Side-Scan Sonar

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Mao, Weibo Zhao, Dongdong Chen, Peng Liu, Jian Liang, Shihui Shen, Yi Guo, Xinxin Liang, Ronghua Zhejiang Univ Technol Coll Comp Sci & Technol Hangzhou 310023 Zhejiang Peoples R China Chinese Acad Sci Inst Deep Sea Sci & Engn Sanya 572000 Hainan Peoples R China

Side-scan sonar technology plays an important role in seabed topographic exploration and mapping. However, the mainstream matched filtering technique does not fully utilize the bandwidth and the distance matching is susceptible to the Doppler effect, which limits the imaging effect and towing speed. In addition, the synthetic aperture method of radar is difficult to be adapted to sonar due to the transmission characteristics of acoustic waves in water. To address these issues, we propose a lateral multi-beam tiling synthetic aperture algorithm framework for linear array side-scan sonar (LMBT-SAS). The core idea is to improve long-range information acquisition capabilities through beam focusing, and indirectly increase the azimuth pulse repetition rate to serve synthetic aperture processing to increase towing speed. To further improve image performance and towing speed, within this framework, we also proposed some specific technical means. Including short-time Fourier frequency domain beamforming with speed compensation, beam tiling with pseudo pulse repetition, distance-dependent beam weight allocation mechanism and two-stage migration correction. The simulation results and lake test data imaging results demonstrate the effectiveness of the proposed LMBT-SAS algorithm, with a performance of 250 meters of transverse scanning distance and 12 knots of towing speed at 400 KHz signal.

关键词： Synthetic aperture sonar Apertures signal processing algorithms Side-scan sonar Frequency-domain analysis Broadband communication Sonar applications Focusing Azimuth Array signal processing High-speed side-scan sonar multibeam synthetic aperture pseudo-pulse repetition beam tiling short-time Fourier beamforming

来源：评论

学校读者我要写书评

暂无评论

Robust Airborne Radar Clutter Suppression Algorithm via Deep Neural Networks Under Small-Sample Conditions

引用

IEEE SENSORS JOURNAL 2025年第7期25卷 11587-11600页

作者： Huang, Weijun Wang, Tong Cui, Weichen Xidian Univ Natl Lab Radar Signal Proc Xian 710071 Peoples R China Beijing Aerosp Automat Control Inst Beijing 100854 Peoples R China Natl Key Lab Sci & Technol Aerosp Intelligence Con Beijing 100854 Peoples R China

Space-time adaptive processing (STAP) is a powerful technique for clutter suppression in airborne radar systems. However, in practical applications, STAP performance is often compromised by the limited availability of independent and identically distributed (i.i.d.) samples, which restricts its clutter suppression capabilities. Inspired by compressed sensing, researchers have extensively investigated sparse recovery-based STAP (SR-STAP) algorithms, which can achieve near-optimal performance under ideal conditions. Yet, these algorithms experience notable performance degradation in the presence of spatial and temporal errors. To address this issue, we propose a deep neural network framework for clutter suppression, cascading a gridless sparse recovery network (GLSRNet) with a generative adversarial network (GAN), ensuring accurate clutter covariance matrix estimation under small-sample conditions with errors. During GAN training, we observed that the traditional binary cross-entropy (BCE) loss function led to significant oscillations in the training loss, impeding effective convergence. To resolve this, we design a new loss function that achieves stable convergence. Extensive experiments validate the effectiveness of the proposed algorithm in clutter suppression.

关键词： Clutter Training Vectors Generative adversarial networks signal processing algorithms Generators Airborne radar Minimization Accuracy Radar clutter Clutter suppression deep unfolding generative adversarial network (GAN) space-time adaptive processing (STAP)

来源：评论

学校读者我要写书评

暂无评论

A Multivariable Motion Sensor Embedding an Improved Velocity Estimation Algorithm

引用

IEEE SENSORS JOURNAL 2025年第8期25卷 13840-13849页

作者： Mazzoli, Federico Alghisi, Davide Ferrari, Vittorio Univ Brescia Dept Informat Engn I-25123 Brescia Italy Gefran Spa Dept Res & Dev I-25050 Provaglio Diseo Italy

A multivariable motion sensor is presented that embeds into its onboard microcontroller a tailored algorithm, referred to here as the double-path (DP) algorithm, which estimates velocity in real time from position and acceleration signals simultaneously measured by the sensor itself. The multivariable motion sensor consists of a contactless magnetic linear position digital sensor and a triaxial digital accelerometer. The proposed algorithm estimates velocity by suitably mixing the integration of the acceleration and the linear fitting of the position, and it can operate under both trapezoidal and S-curve motion profiles. The velocity estimation accuracy has been assessed through simulations and experimental tests, which involved performance evaluation and a comparative analysis between the proposed algorithm and a Kalman filter (KF) both embedded into the sensor microcontroller. The experimental results are obtained by operating the sensor with a reference trapezoidal motion profile with a maximum velocity of 50 mm/s. The two root-mean-square estimation errors calculated for the sensor moving at constant acceleration and velocity are 1.32% and 0.58% of the maximum velocity, respectively.

关键词： Sensors Magnetic sensors Accelerometers Motion detection signal processing algorithms Magnetic cores Estimation Velocity measurement Noise Intelligent sensors Multivariable sensors sensor fusion smart sensors velocity estimation

来源：评论

学校读者我要写书评

暂无评论

Bayes Optimal Learning in High-Dimensional Linear Regression With Network Side Information

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2025年第1期71卷 565-591页

作者： Nandy, Sagnik Sen, Subhabrata Univ Chicago Dept Stat Chicago IL 60637 USA Harvard Univ Dept Stat Cambridge MA 02138 USA

Supervised learning problems with side information in the form of a network arise frequently in applications in genomics, proteomics and neuroscience. For example, in genetic applications, the network side information can accurately capture background biological information on the intricate relations among the relevant genes. In this paper, we initiate a study of Bayes optimal learning in high-dimensional linear regression with network side information. To this end, we first introduce a simple generative model (called the Reg-Graph model) which posits a joint distribution for the supervised data and the observed network through a common set of latent parameters. Next, we introduce an iterative algorithm based on Approximate Message Passing (AMP) which is provably Bayes optimal under very general conditions. In addition, we characterize the limiting mutual information between the latent signal and the data observed, and thus precisely quantify the statistical impact of the network side information. Finally, supporting numerical experiments suggest that the introduced algorithm has excellent performance in finite samples.

关键词： Bayes methods Message passing Mutual information signal processing algorithms Numerical models Limiting Data models Approximation algorithms Slabs Iterative methods Linear regression random networks approximate message passing mutual information variable selection

来源：评论

学校读者我要写书评

暂无评论

CUSE-TD Algorithm for Precision Detection of Coherent and Uncorrelated signals

引用

IEEE ACCESS 2025年 13卷 11307-11319页

作者： Dakulagi, Rohini Raushan, Ravi Dutta, Soham Alosaimi, Mohammed Theeb Villagomez-Galindo, Miguel Natl Inst Technol Karnataka Dept Elect & Elect Engn Mangaluru 575025 Karnataka India Manipal Acad Higher Educ Dept Elect & Elect Engn Manipal Inst Technol Manipal 576104 Karnataka India King Saud bin Abdulaziz Univ Hlth Sci Coll Sci & Hlth Profess Riyadh 11426 Saudi Arabia Minist Natl Guard Hlth Affairs King Abdullah Int Med Res Ctr Riyadh 11481 Saudi Arabia Univ Michoacana Mech Engn Fac Morelia 58000 Michoacan Mexico

This paper presents a novel algorithm, Coherent and Uncorrelated Sources Estimation using Tensor Decomposition (CUSE-TD), designed to enhance the accuracy of Direction of Arrival (DOA) estimation in Multiple-Input Multiple-Output (MIMO) systems. The CUSE-TD algorithm addresses the complex challenge of estimating both coherent and uncorrelated sources by leveraging advanced sensor-driven tensor decomposition techniques, with an emphasis on Tucker decomposition. By meticulously analyzing factor matrices derived from sensor data, the algorithm effectively decodes intricate spatiotemporal patterns within the received tensor, offering precise source estimation even in scenarios where the number and nature of the sources are unknown. The paper provides an in-depth assessment of the algorithm's performance, highlighting its computational efficiency and ability to handle diverse sensor-driven scenarios. Extensive simulations in intricate sensing environments demonstrate the algorithm's robustness, affirming its potential to significantly improve DOA estimation accuracy in MIMO systems across a wide range of operational conditions. The usefulness of the CUSE-TD algorithm lies in its applicability CUSE -TD to real-world MIMO systems, particularly in complex environments such as radar, wireless communications, and surveillance systems, where accurate source detection is critical for operational effectiveness. Its adaptability to varying sensor configurations and source conditions makes it a powerful tool for enhancing overall system reliability and performance.

关键词： Direction-of-arrival estimation Tensors Estimation signal processing algorithms Accuracy MIMO radar Array signal processing Matrix decomposition signal resolution Radar CUSE-TD algorithm direction of arrival (DOA) estimation multiple-input multiple-output (MIMO) systems tensor decomposition Tucker decomposition coherent sources uncorrelated sources spatiotemporal sensor data computational complexity sensor arrays

来源：评论

学校读者我要写书评

暂无评论

APG-Net: APG Optimization-Induced Multiscale Attention Network for MRI Reconstruction From Undersampled k-Space Measurements

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Gan, Hongping Zhao, Zitong Chen, Yunong Shan, Shanshan Shi, Baoshun Liu, Chunyi Northwestern Polytech Univ Sch Software Xian 710072 Peoples R China Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430072 Peoples R China Soochow Univ Sch Radiol & Interdisciplinary Sci State Key Lab Radiat Med & Protect Suzhou 215123 Peoples R China Yanshan Univ Sch Informat Sci & Engn Qinhuangdao 066004 Peoples R China

Compressed sensing magnetic resonance imaging (CS-MRI) can quickly reconstruct magnetic resonance (MR) images from undersampled k-space measurements, thereby accelerating MRI reconstruction. However, existing CS-MRI methods suffer from insufficient information interaction and an inability to adequately capture image features during the iterative reconstruction process, resulting in the reconstructed images having room for improvement in detail. Inspired by the accelerated proximal gradient (APG) algorithm, we propose an APG optimization-induced deep multiscale attention network, called APG-Net, for accelerating MRI reconstruction from k-space measurements. APG-Net transforms the traditional APG algorithm into a learnable network structure while introducing efficient feature extraction modules to effectively capture the features of the k-space data, resulting in high-quality MR image reconstruction. Specifically, it unrolls the iterative steps of the APG algorithm into a cascade block of fixed reconstruction phases, which include the iterative unfolding module (IUM) and multiscale dual-attention block (MDAB). The proposed IUM enhances feature fusion by leveraging information interaction between iterative phases. Moreover, the customized MDAB is utilized to extract multiscale feature information from the k-space measurements while enhancing attention to spatial and channel-related features in both dimensions, effectively boosting feature extraction capability. Extensive experiments demonstrate that our proposed APG-Net achieves remarkable performance on MRI reconstruction tasks in contrast to state-of-the-art methods. Our code is available at APG-Net.

关键词： Image reconstruction Feature extraction Magnetic resonance imaging Iterative algorithms Data mining signal processing algorithms Reconstruction algorithms Convergence Training Optimization Accelerated proximal gradient (APG) compressed sensing (CS) deep learning (DL) magnetic resonance (MR) imaging undersampled measurements

来源：评论

学校读者我要写书评

暂无评论

Distributed Continuous-Time Optimization With Uncertain Time-Varying Quadratic Cost Functions

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2025年第2期55卷 1526-1536页

作者： Jiang, Liangze Wu, Zheng-Guang Wang, Lei Zhejiang Univ Coll Control Sci & Engn Hangzhou 310027 Peoples R China

This article studies distributed continuous-time optimization for time-varying quadratic cost functions with uncertain parameters. We first propose a centralized adaptive optimization algorithm using partial information of the cost function. It can be seen that even if there are uncertain parameters in the cost function, exact optimization can still be achieved. To solve this problem in a distributed manner when different local cost functions have identical Hessians, we propose a novel distributed algorithm that cascades the fixed-time average estimator and the distributed optimizer. We remove the requirement for the upper bounds of certain complex functions by integrating state-based gains in the proposed design. We further extend this result to address the distributed optimization where the time-varying cost functions have nonidentical Hessians. We prove the convergence of all the proposed algorithms in the global sense. Numerical examples verify the proposed algorithms.

关键词： Cost function Heuristic algorithms Costs Time-varying systems signal processing algorithms Prediction algorithms Vectors Information exchange Eigenvalues and eigenfunctions Upper bound Adaptive gain continuous-time optimization distributed optimization quadratic function time-varying cost function uncertainties

来源：评论

学校读者我要写书评

暂无评论

Distributed Multi-Kernel Maximum Correntropy State-Constrained Kalman Filter Under Deception Attacks

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2025年第1期12卷 533-546页

作者： Wang, Guoqing Zhu, Zhaolei Yang, Chunyu Ma, Lei Dai, Wei Chen, Xinkai China Univ Min & Technol Sch Informat & Control Engn Xuzhou 221116 Peoples R China Shibaura Inst Technol Dept Elect & Informat Syst Saitama 3378570 Japan

In this paper, we investigate the distributed robust state estimation of non-Gaussian systems under unknown deception attacks with the imprecise constraint information. Leveraging the advantage of multi-kernel maximum correntropy criterion (MK-MCC) in non-Gaussian signal processing, a novel maximum-a-posterior like utility function (MAP-LUF) is designed inspired by the traditional 2-norm form cost function, where the inaccurate constraint information is taken into consideration. The direct solution of MAP-LUF gives rise to the centralized MK-MCC based state-constrained Kalman filter (C-MKMCSCKF) through fixed point iteration. Subsequently, the corresponding distributed algorithm is obtained by incorporating the consensus average in the computation of sum terms existing in the C-MKMCSCKF algorithm, which enables local information sharing to approximate the centralized estimation accuracy. Furthermore, we also establish the connection between the proposed centralized algorithm and the Banach theorem through dimension extension, and provide the convergence condition. The effectiveness of our proposed algorithms is validated through comparisons with related works in typical target tracking scenarios over sensor network.

关键词： Noise State estimation Vectors signal processing algorithms Cost function Accuracy Target tracking Kernel Kalman filters Convergence Deception attacks distributed Kalman filter multi-kernel correntropy robust state estimation sensor networks state constraint

来源：评论

学校读者我要写书评

暂无评论

A New Robot Navigation Algorithm Based on Bi-Directional Collaborative Ant Colony Optimization

引用

IEEE SENSORS JOURNAL 2025年第8期25卷 14295-14306页

作者： Chen, Xufei Hu, Yaowei Tang, Pingping Zhang, Hui Jin, Jiong Mao, Shiwen Anhui Normal Univ Sch Phys & Elect Informat Wuhu 241002 Peoples R China Nanjing Univ Posts & Telecommun Sch Commun & Informat Engn Nanjing 210003 Peoples R China Swinburne Univ Technol Sch Engn Melbourne Vic 3122 Australia Auburn Univ Dept Elect & Comp Engn Auburn AL 36849 USA

Swarm intelligence has been widely adopted and successfully applied in the field of autonomous robot navigation. Among various swarm intelligence algorithms, ant colony optimization (ACO) has shown significant potential in addressing complex navigation challenges. However, ACO faces challenges, for example, unclear initial search direction, slow convergence, limited ant flexibility, and the need to simplify robot motion control. To address these challenges, this article presents a novel bi-directional collaborative ACO (BC-ACO) algorithm with key innovations. First, the algorithm adopts a bi-directional ant colony with forward and reverse populations, achieving effective route planning through collaborative decision-making. Second, the algorithm employs an adaptive step-size strategy and a stage-based exploration ant quantity adjustment method. These innovations optimize the balance between exploration and exploitation, accelerate convergence, and address the inefficiencies of traditional ACO methods. Additionally, this article improves the heuristic function by integrating a node distance index within the bi-directional ant colony, guiding the transition of ants between nodes and further accelerating convergence. Simulation results show that BC-ACO reduces the computation time by 73.97% and improves the convergence stability by 63.64% compared to standard ACO. Additionally, BC-ACO successfully plans optimal paths, achieving a 20.85% reduction in path length. Further integration of a local quadratic segmented B-spline (LQ-S B-spline) curve results in paths with smooth transitions, reducing robot motion complexity. In summary, BC-ACO achieves fast global convergence, high stability, and short computation time.

关键词： Heuristic algorithms Bidirectional control Convergence Splines (mathematics) Robots Optimization Navigation Indexes signal processing algorithms Collaboration Bi-directional ant colony collaborative decision-making node distance index segmented B-spline stage-based exploration ant quantity adjustment

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

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

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：