检索结果-内蒙古大学图书馆

Algorithm Unrolling Network With Learnable Sparse Regularization for Interpretable Mechanical Anomaly Detection

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2025年第5期21卷 3786-3795页

作者： Chen, Xiyue Wang, Shibin Wang, Shi-ao Ding, Baoqing Yan, Ruqiang Chen, Xuefeng Xi An Jiao Tong Univ Natl Key Lab Aerosp Power Syst & Plasma Technol Xian 710049 Peoples R China Xi An Jiao Tong Univ Sch Mech Engn Xian 710049 Peoples R China

Sparse representation-based interpretable algorithm unrolling is one of promising techniques for mechanical anomaly detection. In order to enhance the learning and representation capabilities of the algorithm unrolling model, this article proposes a learnable sparse regularization network (LSR-Net). Instead of imposing explicit regularization constraints on the encoding and dictionary during modeling, two subnetworks are designed to learn prior information: $\mathbf{Net}_{\mathbf{X}}$ and $\mathbf{Net}_{\mathbf{D}}$. The model is solved using the half quadratic splitting algorithm, and further unrolls the process of iterative computation into the form of a network. The architecture for encoding learning is structured as input convex neural networks, ensuring LSR-Net can learn meaningful encoding priors. Through the analysis of simulated and experimental data, it has been demonstrated that LSR-Net has strong feature extraction and noise resistance capabilities, and the design of its prior learning architecture is both reasonable and effective. In addition, the visualization of LSR-Net's overall reconstruction and the reconstruction of different dictionary atoms allows for both global and local interpretation of the learning results, thereby providing post hoc interpretability to LSR-Net. The visualization results confirm that LSR-Net is capable of learning features that align with mechanical vibration characteristics.

关键词： Convolution Encoding Feature extraction Anomaly detection Dictionaries Convolutional codes Analytical models Predictive models Neural networks signal processing algorithms Algorithm unrolling anomaly detection convolutional dictionary learning (CDL) interpretable neural network

来源：评论

学校读者我要写书评

暂无评论

Quantile Learn-Then-Test: Quantile-Based Risk Control for Hyperparameter Optimization

引用

IEEE signal processing LETTERS 2024年 31卷 3044-3048页

作者： Farzaneh, Amirmohammad Park, Sangwoo Simeone, Osvaldo Kings Coll London Ctr Intelligent Informat Proc Syst Dept Engn Kings Commun Learning & Informat Proc London WC2R 2LS England

The increasing adoption of Artificial Intelligence (AI) in engineering problems calls for the development of calibration methods capable of offering robust statistical reliability guarantees. The calibration of black box AI models is carried out via the optimization of hyperparameters dictating architecture, optimization, and/or inference configuration. Prior work has introduced learn-then-test (LTT), a calibration procedure for hyperparameter optimization (HPO) that provides statistical guarantees on average performance measures. Recognizing the importance of controlling risk-aware objectives in engineering contexts, this work introduces a variant of LTT that is designed to provide statistical guarantees on quantiles of a risk measure. We illustrate the practical advantages of this approach by applying the proposed algorithm to a radio access scheduling problem.

关键词： Reliability Calibration Artificial intelligence Testing Optimization Wireless communication signal processing algorithms Resource management Hyperparameter optimization Delays hyperparameter optimization risk control multiple hypothesis testing quantiles

来源：评论

学校读者我要写书评

暂无评论

Joint Sparse Bayesian Learning for Channel Estimation in ISAC

引用

IEEE COMMUNICATIONS LETTERS 2024年第8期28卷 1825-1829页

作者： Chen, Kangjian Qi, Chenhao Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Peoples R China

In this letter, we investigate channel estimation in integrated sensing and communication and propose a joint sparse Bayesian learning (JSBL) algorithm for the estimation of sensing and communication (S&C) channels. To capture the joint sparsity between S&C channels arising from the commonalities between the sensing targets and communication scatterers, we conceive an adaptive pattern-coupled prior for S&C channel coefficients, which allows us to formulate the joint S&C channel estimation as a maximum a posteriori (MAP) problem. Then, a tailored expectation-maximization (EM) method including an E-step and an M-step is developed to solve the MAP problem. Specifically, in the E-step, we derive the sparse Bayesian estimates for S&C channels, while in the M-step, we dynamically update the off-grid channel angles and prior parameters. Simulation results demonstrate that the proposed JSBL algorithm can achieve improved S&C channel estimation performance over existing methods due to the exploitation of the joint sparsity between S&C channels.

关键词： Channel estimation Bayes methods Uplink Estimation Vectors signal processing algorithms Linear antenna arrays integrated sensing and communication (ISAC) joint sparsity sparse Bayesian learning

来源：评论

学校读者我要写书评

暂无评论

High-Performance QC-LDPC Layered Decoder Based on Shortcut Updating

引用

IEEE signal processing LETTERS 2024年 31卷 596-600页

作者： Wang, Zhongyong Xie, Zhaoyan Gong, Kexian Yan, Jin Chen, Lianghui Zhengzhou Univ Sch Elect & Informat Engn Zhengzhou 450001 Peoples R China

Quasi-cyclic low density parity check (QC-LDPC) is widely used in various modern communication standards due to its excellent performance and convenient hardware implementation, but its commonly used pipelined layered decoder faces update conflicts that lead to performance degradation. This letter proposes a shortcut-based decoder to mitigate this problem. When the conflicts occur, we avoid the read/write operations of the log-likelihood ratios (LLRs) and abandon the use of the barrel shifters, and feed those LLRs directly to the next layer. we also attempt to utilize the latest messages for inevitable conflicts. Moreover, the proposed architecture significantly reduces the possibility of conflicts by reordering the base graph matrix (BGM) with a shorter update interval of LLRs, which can achieve maximum utilization in the updated messages. Compared to the state-of-the-art approach in the case of DVB-S2, the proposed architecture can deliver up to 0.25 dB performance gain with faster convergence.

关键词： Decoding Parity check codes Phase change materials Random access memory Schedules signal processing algorithms Pipelines Layered decoder pipeline conflicts quasi-cyclic LDPC shortcut updating

来源：评论

学校读者我要写书评

暂无评论

Intelligent Radio signal processing: A Survey

引用

IEEE ACCESS 2021年 9卷 83818-83850页

作者： Quoc-Viet Pham Nhan Thanh Nguyen Thien Huynh-The Le, Long Bao Lee, Kyungchun Hwang, Won-Joo Pusan Natl Univ Korean Southeast Ctr 4th Ind Revolut Leader Educ Busan 46241 South Korea Univ Oulu Ctr Wireless Commun Oulu 90014 Finland Kumoh Natl Inst Technol ICT Convergence Res Ctr Gumi 39177 South Korea Univ Quebec Inst Natl Rech Sci INRS Montreal PQ H5A 1K6 Canada Seoul Natl Univ Sci & Technol Dept Elect & Informat Engn Seoul 01811 South Korea Pusan Natl Univ Dept Biomed Convergence Engn Yangsan 50612 South Korea

Intelligent signal processing for wireless communications is a vital task in modern wireless systems, but it faces new challenges because of network heterogeneity, diverse service requirements, a massive number of connections, and various radio characteristics. Owing to recent advancements in big data and computing technologies, artificial intelligence (AI) has become a useful tool for radio signal processing and has enabled the realization of intelligent radio signal processing. This survey covers four intelligent signal processing topics for the wireless physical layer, including modulation classification, signal detection, beamforming, and channel estimation. In particular, each theme is presented in a dedicated section, starting with the most fundamental principles, followed by a review of up-to-date studies and a summary. To provide the necessary background, we first present a brief overview of AI techniques such as machine learning, deep learning, and federated learning. Finally, we highlight a number of research challenges and future directions in the area of intelligent radio signal processing. We expect this survey to be a good source of information for anyone interested in intelligent radio signal processing, and the perspectives we provide therein will stimulate many more novel ideas and contributions in the future.

关键词： Artificial intelligence signal processing Wireless communication signal processing algorithms Channel estimation signal detection Modulation Artificial intelligence beamforming channel estimation deep learning federated learning machine learning modulation classification radio frequency signal processing

来源：评论

学校读者我要写书评

暂无评论

Distributed Fusion of Labeled Multi-Bernoulli Filters Based on Arithmetic Average

引用

IEEE signal processing LETTERS 2024年 31卷 656-660页

作者： Wei, Jingxin Luo, Feng Qi, Jiawei Luo, Xianxian Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China Xidian Univ Hangzhou Inst Technol Hangzhou 311231 Peoples R China

This letter considers the distributed fusion for Labeled Multi-Bernoulli (LMB) filters under the multi-sensor multi-target tracking scenario. In practice, a novel fusion method that combines the label-free strategy with the fusion method based on the Arithmetic Average (AA) is proposed. Firstly, the label-free version of the LMB posterior is obtained. Then, the corresponding Probability Hypothesis Density (PHD) is acquired and fused by the merging-based PHD-AA fusion method. Finally, the labels of the fused distributions are reassigned. The simulation shows that the distributed LMB filter using the proposed fusion method can achieve more accurate multi-target state estimates with relatively low computational burden under clutter and missed detection environment compared to other existing methods.

关键词： Radio frequency Merging signal processing algorithms Target tracking Partitioning algorithms Radar signal processing Mathematical models Arithmetic average fusion distributed multi-target tracking labeled multi-Bernoulli (LMB) filter random finite set

来源：评论

学校读者我要写书评

暂无评论

A Robust Proportionate Graph Recursive Least Squares Algorithm for Adaptive Graph signal Recovery

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2024年第7期71卷 3608-3612页

作者： Sadigh, Alireza Naeimi Zayyani, Hadi Korki, Mehdi Semnan Univ Dept Comp Sci Semnan *** Iran Qom Univ Technol Dept Elect & Comp Engn Qom Iran Swinburne Univ Technol Sch Sci Comp & Engn Technol Melbourne Vic 3122 Australia

In this brief, we propose a robust proportionate Recursive Least Square (RLS) algorithm to address the problem of adaptive graph signal recovery in the presence of impulsive noise. In this problem, a graph signal should be recovered from only a subset of sampled graph signal in an adaptive manner. To this end, first, the classical graph RLS algorithm in the literature is written in a standard way, which is represented by an error correction term. Second, the proportionate graph RLS is formulated by adding a gain matrix in the recursion. Then, to calculate the gain matrix, an optimization problem with non-negative and scaling constraints is suggested. The active-set approach is utilized to handle these constraints. To enhance robustness against impulsive noise, we utilize optimized coefficients derived from the minimum disturbance principle. Moreover, an analytical lower bound for the minimum disturbance and the computational complexity analysis are given. Simulation results show the advantages of the proposed proportionate-based RLS algorithm over other approaches in terms of the rate of convergence in presence of impulsive noise.

关键词： signal processing algorithms Symmetric matrices Optimization signal processing Laplace equations Convergence Computational complexity Graph signal processing proportionate robust recursive least square adaptive

来源：评论

学校读者我要写书评

暂无评论

A Robust DOA Estimator Based on Maximum Mixture Complex Correntropy for Noisy Input and Impulsive Noise

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2024年第7期71卷 3588-3592页

作者： Zhao, Haiquan Luo, Wenjing Xu, Wenjing Southwest Jiaotong Univ Key Lab Magnet Suspens Technol & Maglev Vehicle Minist Educ Chengdu 610031 Peoples R China Southwest Jiaotong Univ Sch Elect Engn Chengdu 610031 Peoples R China

Inspired by the robustness of maximum mixture complex correntropy (MMCC) to non-Gaussian and impulsive noises, an adaptive nulling antenna structure based MMCC adaptive filtering algorithm is presented in this brief to well solve the problem of direction of arrival (DOA) estimation in the presence of impulsive noise. To eliminate the estimation bias caused by array input noise, a bias compensated MMCC (BCMMCC) algorithm is further proposed. Then, the mean stability and computational complexity of the BCMMCC algorithm are analyzed. Finally, the simulation results show that the proposed BCMMCC algorithm has superior DOA estimation performance in noisy input and impulsive noise environments compared with other adaptive algorithms.

关键词： signal processing algorithms Estimation Direction-of-arrival estimation Inference algorithms Computational complexity Classification algorithms signal resolution Direction of arrival estimation maximum mixture complex correntropy noisy input bias-compensated impulsive noise

来源：评论

学校读者我要写书评

暂无评论

Deep Representation Learning for Affective Speech signal Analysis and processing: Preventing unwanted signal disparities

引用

IEEE signal processing MAGAZINE 2021年第6期38卷 22-38页

作者： Lee, Chi-Chun Sridhar, Kusha Li, Jeng-Lin Lin, Wei-Cheng Su, Bo-Hao Busso, Carlos Natl Tsing Hua Univ Elect Engn Dept Hsinchu 300044 Taiwan Univ Texas Dallas Dallas TX 75080 USA Natl Tsing Hua Univ Hsinchu 300044 Taiwan Univ Texas Dallas Elect & Comp Engn Dallas TX 75080 USA Natl Tsing Hua Univ Dept Elect Engn Hsinchu 300044 Taiwan Univ Texas Dallas Elect Engn Dept Dallas TX 75080 USA

Speech emotion recognition (SER) is an important research area, with direct impacts in applications of our daily lives, spanning education, health care, security and defense, entertainment, and human-computer interaction. The advances in many other speech signal modeling tasks, such as automatic speech recognition, text-to-speech synthesis, and speaker identification, have led to the current proliferation of speech-based technology. Incorporating SER solutions into existing and future systems can take these voice-based solutions to the next level. Speech is a highly nonstationary signal, with dynamically evolving spatial-temporal patterns. It often requires a sophisticated representation modeling framework to develop algorithms capable of handling real-life complexities.

关键词： Deep learning Entertainment industry signal processing algorithms Medical services Complexity theory Security Speech processing Affective computing Heuristic algorithms Emotion recognition signal analysis

来源：评论

学校读者我要写书评

暂无评论

Sparsity-Based Adaptive Beamforming for Coherent signals With Polarized Sensor Arrays

引用

IEEE signal processing LETTERS 2024年 31卷 2415-2419页

作者： Liu, Tianpeng Cheng, Yun Shi, Junpeng Liu, Zhen Liu, Yongxiang Natl Univ Def Technol Coll Elect Sci & Technol Changsha 410073 Peoples R China Natl Univ Def Technol Coll Elect Engn Hefei 230037 Peoples R China

A sparsity-based adaptive beamforming (ABF) method is introduced to effectively process coherent signals with polarized sensor arrays (PSA). This method exploits the spatial sparsity of observed signals by transforming it into row-sparsity within a waveform-polarization composite matrix through data reorganization. This row-sparsity is subsequently cast as an l(2,1) norm minimization problem, characterized by a gridless and compact mathematical expression with a Hermitian Toeplitz matrix. Then, a matrix factorization-based gradient descent (GD) algorithm is introduced to effectively resolve this optimization problem. The experimental evaluations demonstrate that the GD algorithm significantly outperforms the MOSEK solver in terms of computational efficiency. Further comparative analysis demonstrates that the proposed method outperforms the existing techniques, especially in contexts of low signal-to-noise ratio (SNR), with a moderate increase in computational runtime.

关键词： Vectors Array signal processing signal to noise ratio Sensor arrays Covariance matrices signal processing algorithms Adaptive arrays Adaptive beamforming coherent signal gradient descent polarized sensor array

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：