检索结果-内蒙古大学图书馆

KFA: Keyword Feature Augmentation for Open Set Keyword Spotting

IEEE signal processing LETTERS 2024年 31卷 2985-2989页

作者： Ko, Kyungdeuk Lee, Bokyeung Hong, Jonghwan Ko, Hanseok Korea Univ Seoul 02841 South Korea

In recent years, with the advancement of deep learning technology and the emergence of smart devices, there has been a growing interest in keyword spotting (KWS), which is used to activate AI systems with automatic speech recognition and text-to-speech. However, smart devices with KWS often encounter false alarm errors when inputting unexpected words. To address this issue, existing KWS methods typically train non-target words as an unknown class. Despite these efforts, there is still a possibility that unseen words not trained as part of the unknown class could be misclassified as one of the target words. To overcome this limitation, we propose a new method named Keyword Feature Augmentation (KFA) for open-set KWS. KFA performs feature augmentation through adversarial learning to increase the loss. The augmented features are constrained within a limited space using label smoothing. Unlike other generative model-based open set recognition (OSR) methods, KFA does not require any additional training parameters or repeated operation for inference. As a result, KFA has achieved a 0.955 AUROC score and 97.34% target class accuracy for Google Speech Commands V1, and a 0.959 AUROC score and 98.17% target class accuracy for Google Speech Commands V2, which is the highest performance when compared to various OSR methods.

关键词： Training Internet Convolution Artificial intelligence signal processing algorithms Pattern classification Inference algorithms Generative adversarial networks Classification algorithms Accuracy Deep learning keyword spotting open set recognition

来源：评论

学校读者我要写书评

暂无评论

Detecting Anxiety via Machine Learning algorithms: A Literature Review

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2025年

作者： Tayarani-N., M. -H. Shahid, Shamim Ibne Univ Hertfordshire Sch Phys Engn & Comp Sci HATFIELD Hatfield AL10 9AB England

Recent machine learning (ML) advances have opened up new possibilities for addressing various challenges. Given their ability to tackle complex problems, the use of ML algorithms in diagnosing mental health disorders has seen substantial growth in both the number and scope of studies. Anxiety, a major health concern in today's world, affects a significant portion of the population. Individuals with anxiety often exhibit distinct characteristics compared to those without the disorder. These differences can be observed in their outward appearance-such as voice, facial expressions, gestures, and movements-and in less visible factors like heart rate, blood test results, and brain imaging data. In this context, numerous studies have utilized ML algorithms to extract a diverse range of features from individuals with anxiety, aiming to build predictive models capable of accurately identifying those affected by the disorder. This paper performs a comprehensive literature review on the state-of-the-art studies that employ machine learning algorithms to identify anxiety. This paper aims to cover a wide range of studies and categorize them based on their methodologies and data types used.

关键词： Anxiety disorders Machine learning algorithms Mental health Prediction algorithms Feature extraction Depression Accuracy Systematic literature review Sleep signal processing algorithms Anxiety disorder machine learning artificial intelligence mental disorder social signal processing affective computing

来源：评论

学校读者我要写书评

暂无评论

A Low Complexity Linear Precoding Method for Extremely Large-Scale MIMO Systems

IEEE OPEN JOURNAL OF VEHICULAR TECHNOLOGY

引用

IEEE OPEN JOURNAL OF VEHICULAR TECHNOLOGY 2025年 6卷 240-255页

作者： Berra, Salah Benchabane, Abderrazak Chakraborty, Sourav Maruta, Kazuki Dinis, Rui Beko, Marko Kasdi Merbah Ouargla Univ Dept Elect & Telecommun Lab LAGE Ouargla 30000 Algeria Univ Lusofona COPELABS P-1749024 Lisbon Portugal Cooch Behar Govt Engn Coll Dept Elect & Commun Engn Cooch Behar 736170 India Tokyo Univ Sci Fac Engn Dept Elect Engn Tokyo 1628601 Japan FCT UNL Inst Telecomunicacoes P-2825515 Caparica Portugal Univ Lisbon Inst Super Tecn P-1649004 Lisbon Portugal

Massive multiple-input multiple-output (MIMO) systems are critical technologies for the next generation of networks. In this field of research, new forms of deployment are emerging, such as extremely large-scale MIMO (XL-MIMO), in which the antenna array at the base station (BS) is of extreme dimensions. As a result, spatial non-stationary features emerge as users view just a section of the antenna array, known as the visibility regions (VRs). The XL-MIMO systems can achieve higher spectral efficiency, improve cell coverage, and provide significantly higher data rates than standard MIMO systems. It is a promising technology for future sixth-generation (6G) networks. However, due to the large number of antennas, linear precoding algorithms such as Zero-Forcing (ZF) and regularized Zero-Forcing (RZF) methods suffer from unacceptable computational complexity, primarily due to the required matrix inversion. This work aims to develop low-complexity precoding techniques for the downlink XL-MIMO system. These low-complexity linear precoding methods are based on Gauss-Seidel (GS) and Successive Over-Relaxation (SOR) techniques, which avoid calculating the complex matrix inversion and lead to stable linear precoding performance. To further enhance linear precoding performance, we incorporate the Chebyshev acceleration method with the SOR and GS methods, referred to as the Cheby-SOR and Cheby-GS methods. As these proposed methods require optimizing parameters, we create a deep unfolded network (DUN) to optimize the algorithm parameters. Our performance results demonstrate that the proposed method significantly reduces computational complexity from to O(K-2), where $K$ represents the number of users. Moreover, our approach outperforms the original algorithms, requiring only a few iterations to achieve the RZF bit error rate (BER) performance.

关键词： Precoding Massive MIMO Downlink Iterative methods Computational complexity signal processing algorithms Chebyshev approximation Interference Vehicular and wireless technologies Artificial neural networks XL-MIMO non-stationary linear precoding Chebychev acceleration low-complexity iterative method deep unfolding

来源：评论

学校读者我要写书评

暂无评论

RFI-Aware and Low-Cost Maximum Likelihood Imaging for High-Sensitivity Radio Telescopes

引用

IEEE signal processing LETTERS 2024年 31卷 2960-2964页

作者： Wang, J. El Korso, M. N. Bacharach, L. Larzabal, P. Univ Paris Saclay SATIE F-91190 Gif Sur Yvette France Univ Paris Saclay L2S F-91190 Gif Sur Yvette France

This paper addresses the challenge of interference mitigation and reduction of computational cost in the context of radio interferometric imaging. We propose a novel maximum-likelihood-based methodology based on the antenna sub-array switching technique, which strikes a refined balance between imaging accuracy and computational efficiency. In addition, we tackle robustness regarding radio interference by modeling the additive noise as t-distributed. Through simulation results, we demonstrate the superiority of the t-distributed noise model over the conventional Gaussian noise model in scenarios involving interferences. We evidence that our proposed switching approach yields similar imaging performances with far fewer visibilities compared to the full array configuration, thus, diminishing the computational complexity.

关键词： Noise Imaging Vectors Antennas Computational modeling Antenna measurements Antenna arrays Switches signal processing algorithms Sensors Antenna array processing expectation-maximization algorithm interferometry radio-astronomical imaging

来源：评论

学校读者我要写书评

暂无评论

Fast and Accurate Log-Determinant Approximations

引用

IEEE signal processing LETTERS 2024年 31卷 1520-1524页

作者： Deen, Owen Waller, Colton River Ward, John Paul Univ North Carolina Wilmington Dept Math & Stat Wilmington NC 28403 USA North Carolina A&T State Univ Dept Math & Stat Greensboro NC 27408 USA

We consider the problem of estimating log-determinants of large, sparse, positive definite matrices. A key focus of our algorithm is to reduce computational cost, and it is based on sparse approximate inverses. The algorithm can be implemented to be adaptive, and it uses graph spline approximation to improve accuracy. We illustrate our approach on classes of large sparse matrices.

关键词： Sparse matrices Splines (mathematics) Approximation algorithms Indexes signal processing algorithms Laplace equations Vectors Graph models graph splines log-determinant positive definite sparse matrices

来源：评论

学校读者我要写书评

暂无评论

Maximum Correntropy Kalman Filter for Linear Discrete-Time Systems With Intermittent Observations and Non-Gaussian Noise

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2024年第6期71卷 3246-3250页

作者： Song, Xinmin Zhang, Min Zheng, Wei Xing Liu, Zheng Shandong Normal Univ Sch Informat Sci & Engn Jinan 250358 Shandong Peoples R China Western Sydney Univ Sch Comp Data & Math Sci Sydney NSW 2751 Australia

During data transmission over unreliable communication networks, intermittent observations may appear due to data loss or packet drops. Meanwhile, in practical applications, communication networks are usually disturbed by non-Gaussian noise, e.g., heavy-tailed impulsive noise. To improve the robustness of the Kalman filter with intermittent observations (IOKF) against non-Gaussian noise, this brief proposes the maximum correntropy Kalman filter with intermittent observations (MCIOKF), exploiting only the information arrival probability to design and implement the estimator. The robust maximum correntropy, instead of the conventional minimum mean square error, is taken as the optimality criterion to make the estimator perform better than the IOKF. Similar to the traditional IOKF, the MCIOKF performs time update according to the state estimation and covariance propagation equation. In the measurement updates, the developed MCIOKF adopts a widely used fixed-point algorithm and establishes the augmented model of the IOKF by designing a modified error vector whose covariance matrix contains the state covariance function. Finally, the effectiveness and robustness of the proposed algorithm are validated by a vehicle tracking example.

关键词： Kalman filters Noise measurement Robustness signal processing algorithms Gaussian noise Covariance matrices Protocols Kalman filter intermittent observations maximum correntropy information arrival probability non-Gaussian noise

来源：评论

学校读者我要写书评

暂无评论

Learning-Based Tracking-Before-Detect for Unconstrained Indoor Human Tracking Using RF signal

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2025年第6期24卷 4777-4793页

作者： Wu, Zhi Zhang, Dongheng Shang, Zixin Yuan, Yuqin Gong, Hanqin Wang, Binquan Lu, Zhi Li, Yadong Hu, Yang Sun, Qibin Chen, Yan Univ Sci & Technol China Sch Cyber Sci & Technol Hefei 230026 Peoples R China Univ Washington Dept Elect & Comp Engn Seattle WA 98195 USA

Human tracking plays a crucial role in various wireless sensing applications. However, recent advancements have primarily focused on constrained experimental scenarios with less interference, often involving a few individuals performing actions in an empty space without obstacles. In empirical unconstrained scenarios, such as daily office scenes, severe interference and attenuation caused by chaotic environments is inevitable which results in dramatic performance degradation. In this paper, we introduce TBDNet, which incorporates tracking-before-detect (TBD) from conventional signal processing into learning-based models, achieving impressive tracking performance in unconstrained scenarios. TBDNet follows first-track-then-detect pipeline. It maps input heatmap sequence into high-level frame-wise features to adapt the time-varying intensity distribution and motion pattern of targets. After that, the temporal information is accumulated in feature space to obtain trace proposals. We then predict the accurate positions and probability of traces at each timestamp. To assess the efficiency of TBDNet, we collect and release the first RF-UNIT (RF-based Unconstrained Indoor Tracking) dataset, which comprises 4,030,880 radar heatmaps and the corresponding tracking annotations under 6 different scenarios. To our knowledge, RF-UNIT is the first dataset for RF-based human tracking in unconstrained scenes. We anticipate that TBDNet and the RF-UNIT dataset will significantly contribute to the advancement of RF-based sensing technologies.

关键词： Radar tracking Target tracking Tracking Sensors Radar Interference Radar applications signal processing algorithms Heating systems signal processing Wireless sensing RF-based human tracking tracking-before-detect unconstrained scenarios

来源：评论

学校读者我要写书评

暂无评论

Perception-Based Prediction for Efficient Kinesthetic Coding

引用

IEEE signal processing LETTERS 2024年 31卷 2530-2534页

作者： Xu, Yiwen Huang, Qingfeng Zheng, Quanfei Fang, Ying Zhao, Tiesong Fuzhou Univ Fujian Key Lab Intelligent Proc & Wireless Transmi Fuzhou 350025 Peoples R China Fuzhou Univ Coll Zhicheng Fuzhou 350025 Peoples R China

Integrating haptic feedback with audio and video not only expands the perceptual dimensions of multimedia applications but also enhances user engagement and experience. However, higher signal sampling rates and multi-degree-freedom in haptic interaction increase data significantly. For low-latency and reliable transmission of haptic signal (i.e. tactile and kinesthetic signals), efficient haptic coding is crucial. Existing algorithms overlook haptic signal characteristics, leaving room for improvement. We analyze the statistical characteristics of kinesthetic signals in-depth. Based on the local linear characteristics of position and velocity signals, and the sparse distribution of force signal, we propose an improved kinesthetic coding algorithm by combining dead-zone coding with segmented linear prediction. Extensive experiments on the standard datasets of the IEEE P1918.1.1 Haptic Codecs Task Group demonstrate the superior performance compared to state-of-the-art methods, achieving a more than halved reduction in data transmission rates with high signal-to-noise ratios and structural similarity.

关键词： Encoding Haptic interfaces Prediction algorithms Force signal processing algorithms Mathematical models Linear regression Data compression haptics kinesthetic coding regression algorithm

来源：评论

学校读者我要写书评

暂无评论

Non-Stationary Delayed Combinatorial Semi-Bandit With Causally Related Rewards

IEEE OPEN JOURNAL OF SIGNAL PROCESSING

引用

IEEE OPEN JOURNAL OF signal processing 2025年 6卷 369-384页

作者： Ghoorchian, Saeed Bilaj, Steven Maghsudi, Setareh Ruhr Univ Bochum Fac Elect Engn & Informat Technol D-44801 Bochum Germany Univ Sci & Technol Tehran Iran Univ Darmstadt Darmstadt Germany Tech Univ Berlin Berlin Germany

Sequential decision-making under uncertainty is often associated with long feedback delays. Such delays degrade the performance of the learning agent in identifying a subset of arms with the optimal collective reward in the long run. This problem becomes significantly challenging in a non-stationary environment with structural dependencies amongst the reward distributions associated with the arms. Therefore, besides adapting to delays and environmental changes, learning the causal relations alleviates the adverse effects of feedback delay on the decision-making process. We formalize the described setting as a non-stationary and delayed combinatorial semi-bandit problem with causally related rewards. We model the causal relations by a directed graph in a stationary structural equation model. The agent maximizes the long-term average payoff, defined as a linear function of the base arms' rewards. We develop a policy that learns the structural dependencies from delayed feedback and utilizes that to optimize the decision-making while adapting to drifts. We prove a regret bound for the performance of the proposed algorithm. Besides, we evaluate our method via numerical analysis using synthetic and real-world datasets to detect the regions that contribute the most to the spread of Covid-19 in Italy.

关键词： Delays Decision making Mathematical models Vectors Numerical analysis COVID-19 signal processing algorithms signal processing Uncertainty Synthetic data Combinatorial multi-armed bandit delayed feedback non-stationary environment online learning uncertainty

来源：评论

学校读者我要写书评

暂无评论

Optimised Serial Commutator FFT Architecture in Terms of Multiplexers

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2024年第1期71卷 445-449页

作者： Fang, Hongji Ma, Zhenguo Yu, Feng Zhao, Bei Zhang, Bo Zhejiang Univ Coll Biomed Engn & Instrument Sci Hangzhou 310027 Peoples R China Hangzhou Dianzi Univ Coll Comp Sci & Technol Hangzhou 310005 Peoples R China Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou 310027 Peoples R China

This brief introduces an optimised architecture for a serial commutator (SC) pipelined fast Fourier transform (FFT) with fully real data paths, utilising a novel data management scheme based on bit-dimension permutation circuits. By employing this approach, we minimised multiplexer utilisation by 80% in every processing element (PE) and 50% in every permutation circuit of each stage compared to the conventional SC architecture. Additionally, the proposed structure achieved full hardware utilisation with a natural-order input/output (I/O) datastream. Experimental results demonstrate that our architecture reduced look-up table (LUT) logic by 31% compared to the prior design.

关键词： Computer architecture Multiplexing Indexes Hardware Germanium Discrete Fourier transforms signal processing algorithms FFT pipelined architecture bit-dimension permutations serial commutator

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：