期刊文献+
共找到83,995篇文章
< 1 2 250 >
每页显示 20 50 100
A Comparative Study of Optimized-LSTM Models Using Tree-Structured Parzen Estimator for Traffic Flow Forecasting in Intelligent Transportation 被引量:1
1
作者 Hamza Murad Khan Anwar Khan +3 位作者 Santos Gracia Villar Luis Alonso DzulLopez Abdulaziz Almaleh Abdullah M.Al-Qahtani 《Computers, Materials & Continua》 2025年第5期3369-3388,共20页
Traffic forecasting with high precision aids Intelligent Transport Systems(ITS)in formulating and optimizing traffic management strategies.The algorithms used for tuning the hyperparameters of the deep learning models... Traffic forecasting with high precision aids Intelligent Transport Systems(ITS)in formulating and optimizing traffic management strategies.The algorithms used for tuning the hyperparameters of the deep learning models often have accurate results at the expense of high computational complexity.To address this problem,this paper uses the Tree-structured Parzen Estimator(TPE)to tune the hyperparameters of the Long Short-term Memory(LSTM)deep learning framework.The Tree-structured Parzen Estimator(TPE)uses a probabilistic approach with an adaptive searching mechanism by classifying the objective function values into good and bad samples.This ensures fast convergence in tuning the hyperparameter values in the deep learning model for performing prediction while still maintaining a certain degree of accuracy.It also overcomes the problem of converging to local optima and avoids timeconsuming random search and,therefore,avoids high computational complexity in prediction accuracy.The proposed scheme first performs data smoothing and normalization on the input data,which is then fed to the input of the TPE for tuning the hyperparameters.The traffic data is then input to the LSTM model with tuned parameters to perform the traffic prediction.The three optimizers:Adaptive Moment Estimation(Adam),Root Mean Square Propagation(RMSProp),and Stochastic Gradient Descend with Momentum(SGDM)are also evaluated for accuracy prediction and the best optimizer is then chosen for final traffic prediction in TPE-LSTM model.Simulation results verify the effectiveness of the proposed model in terms of accuracy of prediction over the benchmark schemes. 展开更多
关键词 Short-term traffic prediction sequential time series prediction TPE tree-structured parzen estimator LSTM hyperparameter tuning hybrid prediction model
在线阅读 下载PDF
Distributed predefined-time estimator-based affine formation target-enclosing maneuver control for cooperative underactuated quadrotor UAVs with fault-tolerant capabilities 被引量:1
2
作者 Yang XU Yuanfang QU +2 位作者 Delin LUO Haibin DUAN Zhengyu GUO 《Chinese Journal of Aeronautics》 2025年第1期471-490,共20页
The paper presents a two-layer,disturbance-resistant,and fault-tolerant affine formation maneuver control scheme that accomplishes the surrounding of a dynamic target with multiple underactuated Quadrotor Unmanned Aer... The paper presents a two-layer,disturbance-resistant,and fault-tolerant affine formation maneuver control scheme that accomplishes the surrounding of a dynamic target with multiple underactuated Quadrotor Unmanned Aerial Vehicles(QUAVs).This scheme mainly consists of predefinedtime estimators and fixed-time tracking controllers,with a hybrid Laplacian matrix describing the communication among these QUAVs.At the first layer,we devise predefined time estimators for leading and following QUAVs,enabling accurate estimation of desired information.In the second layer,we initially devise a fixed-time hybrid observer to estimate unknown disturbances and actuator faults.Fixedtime translational tracking controllers are then proposed,and the intermediary control input from these controllers is used to extract the desired attitude and angular velocities for the fixed-time rotational tracking controllers.We employ an exact tracking differentiator to handle variables that are challenging to differentiate directly.The paper includes a demonstration of the control system stability through mathematical proof,as well as the presentation of simulation results and comparative simulations. 展开更多
关键词 Affine formation maneuver control Target tracking Fixed-time control Quadrotor unmanned aerial vehicle Target enclosing Predefined-time estimation
原文传递
Cytokeratin 18 fragment is associated with steatosis-associated fibrosis estimator score and lipid in patients with steatotic liver disease
3
作者 Tatsuki Ichikawa Satoshi Miuma +10 位作者 Mio Yamashima Shinobu Yamamichi Makiko Koike Yusuke Nakano Hiruyuki Yajima Osamu Miyazaki Tomonari Ikeda Takauma Okamura Naohiro Komatsu Miruki Yoshino Hisamitsu Miyaaki 《World Journal of Hepatology》 2025年第11期181-199,共19页
BACKGROUND Serum cytokeratin 18 fragment(CK18F)has been developed as a new noninvasive test(NIT)for risk assessment of steatotic liver disease(SLD);however,there are few reports on its relationship with existing NITs ... BACKGROUND Serum cytokeratin 18 fragment(CK18F)has been developed as a new noninvasive test(NIT)for risk assessment of steatotic liver disease(SLD);however,there are few reports on its relationship with existing NITs and association with cardiometabolic risk factors(CMRFs).AIM To clarify the relationship among CK18F,NITs,and CMRF.METHODS We included 125 patients who were assessed for SLD and had CK18F measured in cross-sectional study.The fibrosis-4 index(FIB-4),steatosis-associated fibrosis estimator(SAFE)score,liver stiffness(LS),controlled attenuation parameter,and FibroScan-aspartate aminotransferase(FAST)score were compared with CK18F as existing NITs.RESULTS CK18F was associated with aspartate aminotransferase,alanine aminotransferase,and triglyceride(TG).FAST and SAFE score were associated with high CK18F(>260 U/L),but not FIB-4 or LS.The cut-off values for TG and high-density lipoprotein(HDL)cholesterol used to determine high CK18F using receiver operating characteristics analysis were 126 mg/dL and 56 mg/dL respectively.High TG(>126 mg/dL)and low HDL(<56 mg/dL)were associated with high CK18F.The risk of high CK18F was higher when high TG and low HDL were combined than when each was present alone.CMRF was higher in the high CK18F group,but was not associated with CK18F levels.However,when the TG and HDL criteria for CMRF were replaced by TG>126 mg/mL and HDL<56 mg/dL,modified CMRF(mCMRF)was associated with CK18F levels,with a higher risk of high CK18F than CMRF.CONCLUSION CK18F is a new NIT associated with SAFE score and FAST.High TG,low HDL,and mCMRF are associated with high CK18F. 展开更多
关键词 Cytokeratin 18 fragment Steatosis-associated fibrosis estimator score Triglyceride High-density lipoprotein cholesterol Steatotic liver disease
暂未订购
Continuous Bayesian probability estimator in predictions of nuclear charge radii
4
作者 Jian Liu Kai-Zhong Tan +4 位作者 Lei Wang Wan-Qing Gao Tian-Shuai Shang Jian Li Chang Xu 《Nuclear Science and Techniques》 2025年第11期283-293,共11页
Recently,machine learning has become a powerful tool for predicting nuclear charge radius RC,providing novel insights into complex physical phenomena.This study employs a continuous Bayesian probability(CBP)estimator ... Recently,machine learning has become a powerful tool for predicting nuclear charge radius RC,providing novel insights into complex physical phenomena.This study employs a continuous Bayesian probability(CBP)estimator and Bayesian model averaging(BMA)to optimize the predictions of RCfrom sophisticated theoretical models.The CBP estimator treats the residual between the theoretical and experimental values of RCas a continuous variable and derives its posterior probability density function(PDF)from Bayesian theory.The BMA method assigns weights to models based on their predictive performance for benchmark nuclei,thereby accounting for the unique strengths of each model.In global optimization,the CBP estimator improved the predictive accuracy of the three theoretical models by approximately 60%.The extrapolation analyses consistently achieved an improvement rate of approximately 45%,demonstrating the robustness of the CBP estimator.Furthermore,the combination of the CBP and BMA methods reduces the standard deviation to below 0.02 fm,effectively reproducing the pronounced shell effects on RCof the Ca and Sr isotope chains.The studies in this paper propose an efficient method to accurately describe RCof unknown nuclei,with potential applications in research on other nuclear properties. 展开更多
关键词 Machine learning Nuclear charge radii Continuous Bayesian probability estimator Bayesian model averaging
在线阅读 下载PDF
A New Class of Biased Linear Estimators in Deficient-rank Linear Models 被引量:1
5
作者 归庆明 段清堂 +1 位作者 周巧云 郭建锋 《Chinese Quarterly Journal of Mathematics》 CSCD 2001年第1期71-78,共8页
In this paper, we define a new class of biased linear estimators of the vector of unknown parameters in the deficient_rank linear model based on the spectral decomposition expression of the best linear minimun bias es... In this paper, we define a new class of biased linear estimators of the vector of unknown parameters in the deficient_rank linear model based on the spectral decomposition expression of the best linear minimun bias estimator. Some important properties are discussed. By appropriate choices of bias parameters, we construct many interested and useful biased linear estimators, which are the extension of ordinary biased linear estimators in the full_rank linear model to the deficient_rank linear model. At last, we give a numerical example in geodetic adjustment. 展开更多
关键词 deficient_rank model best linear minimum bias estimator generalized principal components estimator mean squared error condition number
在线阅读 下载PDF
基于内窥镜视觉的机器人辅助手术中力估计方法
6
作者 邢元 王建敏 +3 位作者 马剑雄 唐吉思 马志康 史靖 《天津大学学报(自然科学与工程技术版)》 北大核心 2026年第1期99-110,共12页
在机器人辅助微创手术中,精确的力反馈对于提高手术操作的安全性与质量至关重要.然而,现有的解决方案在实际应用场景中仍面临着如小型化、精准化和普适性等多重挑战,限制了其在复杂临床场景中的广泛应用.因此,如何实现高精度、低成本且... 在机器人辅助微创手术中,精确的力反馈对于提高手术操作的安全性与质量至关重要.然而,现有的解决方案在实际应用场景中仍面临着如小型化、精准化和普适性等多重挑战,限制了其在复杂临床场景中的广泛应用.因此,如何实现高精度、低成本且适用于多种组织类型的力估计成为研究重点.为此,基于手术机器人配备的内窥镜系统,构建了包含多种材料和丰富力学信息数据集,并提出了结合注意力机制的深度学习模型,以优化内窥镜视觉信息的特征提取,从而提升力估计的准确性和鲁棒性.模型以材料变形的内窥镜图像作为输入,结合卷积神经网络的特征提取能力和循环神经网络的时序建模能力,实现对手术器械与组织之间交互力的精确估计.此外,为进一步探究注意力机制在力估计任务中的作用和优化策略,提出了3种不同的注意力模块引入方案.实验结果表明:采用DenseNet-BiLSTM结构并引入SENet模块的模型在3种材料上取得了最佳性能,显著提升了模型的整体表现;同时,注意力模块的位置对不同组织材料的力估计效果具有差异性影响.研究验证了基于内窥镜视觉的深度学习方法在精确估计手术过程中器械与组织间的交互力方面的有效性和可行性,为未来机器人辅助微创手术系统的发展和优化提供了新的方向与理论依据. 展开更多
关键词 机器人辅助微创手术 视觉反馈 交互力估计 注意力机制 深度学习
在线阅读 下载PDF
基于最佳滑移率估计的汽车EMB防抱死控制
7
作者 潘公宇 熊浩东 《郑州大学学报(工学版)》 北大核心 2026年第1期58-65,共8页
为了解决传统的逻辑门限式ABS控制方法存在无法充分利用路面利用附着系数以及滑移率波动较大的问题,提出了一种基于最佳滑移率估计的汽车EMB防抱死控制策略。首先,建立轮胎滑移率与路面利用附着系数之间的非线性模型;其次,通过一种分段... 为了解决传统的逻辑门限式ABS控制方法存在无法充分利用路面利用附着系数以及滑移率波动较大的问题,提出了一种基于最佳滑移率估计的汽车EMB防抱死控制策略。首先,建立轮胎滑移率与路面利用附着系数之间的非线性模型;其次,通过一种分段式的估计算法来快速准确地跟踪最佳滑移率;最后,基于最佳滑移率的估计结果,设计了积分滑模控制器,通过精确调节EMB制动力矩和电机制动力矩,使前后轮的滑移率维持在各自的最佳滑移率,保证车辆在不同路面条件下的最佳制动距离。仿真结果表明:所采用的估计算法都能够快速准确识别当前路面的最佳滑移率,估计出的最佳滑移率在稳态时与实际的最佳滑移率的最大误差不超过3%,且积分滑模控制器可以精确控制滑移率保持在最佳滑移率附近,与CarSim内置的ABS控制策略相比,单一路面工况制动总时间缩短了10.8%,制动总距离减少了15.8%,对接路面工况制动总时间缩短了18.0%,制动总距离减少了22.2%。 展开更多
关键词 最佳滑移率 电子机械制动 ABS 估计算法 积分滑模控制
在线阅读 下载PDF
ON BAHADUR-TYPE ASYMPTOTIC EFFICIENCY OF POINT ESTIMATORS UNDER IRREGULAR TRUNCATED DISTRIBUTION FAMILY 被引量:1
8
作者 陈桂景 王尧弘 李宁宁 《Acta Mathematica Scientia》 SCIE CSCD 1996年第2期142-152,共11页
In this paper, the optimal convergence rates of point estimators have been found under the irregular truncated distribution family, and corresponding Bahadurtype asymptotic efficiencies have been established. It has b... In this paper, the optimal convergence rates of point estimators have been found under the irregular truncated distribution family, and corresponding Bahadurtype asymptotic efficiencies have been established. It has beed justified that commonly used estimators are all efficient in this sense. 展开更多
关键词 irregular truncated family Bahadnr-type asymptotic efficiency commonly used estimator.
在线阅读 下载PDF
环境治理数字化转型成效评估、动态演进与空间收敛
9
作者 杨恺钧 罗灏飞 《生态经济》 北大核心 2026年第1期179-191,共13页
环境治理数字化转型是推动生态文明建设与治理能力现代化的重要路径,对实现绿色低碳高质量发展具有基础性支撑作用。论文基于中国30个省份2009—2023年的面板数据,采用Super-SBM模型测度环境治理数字化转型效率,并通过核密度估计、马尔... 环境治理数字化转型是推动生态文明建设与治理能力现代化的重要路径,对实现绿色低碳高质量发展具有基础性支撑作用。论文基于中国30个省份2009—2023年的面板数据,采用Super-SBM模型测度环境治理数字化转型效率,并通过核密度估计、马尔科夫链转移概率矩阵及空间收敛模型分析其动态演进与敛散特征。研究发现:(1)我国环境治理数字化转型效率总体呈上升趋势,但区域差异明显,呈现东高西低的核心—边缘结构。(2)各区域动态演进特征差异较大,转型效率存在“俱乐部收敛”现象,实现“跨越式”发展概率较小,且断点回归检验环境治理数字化转型政策对环境治理数字化转型效率具有正向推动作用。(3)全国及各区域环境治理数字化转型效率普遍存在σ收敛、绝对β收敛和条件β收敛特征。此外,产业结构、城镇化水平、科研投入、政府干预、人口密度和对外开放程度对不同区域环境治理数字化转型效率的影响存在显著异质性。 展开更多
关键词 环境治理数字化转型 Super-SBM模型 核密度估计 马尔科夫链 空间收敛模型
原文传递
Robustness of Minimum Norm Quadratic Unbiased Estimator of Variance Under the General Linear Model
10
作者 张宝学 罗季 李馨 《Journal of Beijing Institute of Technology》 EI CAS 2002年第1期97-100,共4页
Necessary and sufficient conditions for equalities between a 2 y′(I-P Xx)y and minimum norm quadratic unbiased estimator of variance under the general linear model, where a 2 is a known positive number, are... Necessary and sufficient conditions for equalities between a 2 y′(I-P Xx)y and minimum norm quadratic unbiased estimator of variance under the general linear model, where a 2 is a known positive number, are derived. Further, when the Gauss? Markov estimators and the ordinary least squares estimator are identical, a relative simply equivalent condition is obtained. At last, this condition is applied to an interesting example. 展开更多
关键词 general linear model orthogonal projector minimum norm quadratic unbiased estimator
在线阅读 下载PDF
Statistical condition assessment of existing structures using virtual work error estimator
11
作者 蔡晶 吴智深 《Journal of Southeast University(English Edition)》 EI CAS 2007年第4期577-582,共6页
A statistical damage detection and condition assessment scheme for existing structures is developed. First a virtual work error estimator is defined to express the discrepancy between a real structure and its analytic... A statistical damage detection and condition assessment scheme for existing structures is developed. First a virtual work error estimator is defined to express the discrepancy between a real structure and its analytical model, with which a system identification algorithm is derived by using the improved Newton method. In order to investigate its properties in the face of measurement errors, the Monte Carlo method is introduced to simulate the measured data. Based on the identified results, their statistical distributions can be assumed, the status of an existing structure can be statistically evaluated by hypothesis tests. A 5-story, two-bay steel frame is used to carry out numerical simulation studies in detail, and the proposed scheme is proved to be effective. 展开更多
关键词 virtual work constitutive parameter parameter estimation hypothesis test
在线阅读 下载PDF
旋翼无人机FMCW SAR成像系统设计
12
作者 周鹏 蒋子仪 邱杭 《电子设计工程》 2026年第1期140-144,149,共6页
为了开发一套低成本且高实用性的旋翼无人机合成孔径雷达(Synthetic Aperture Radar,SAR)成像系统,该文以六旋翼无人机为载体,设计了线性调频连续波(Frequency Modulated Continuous Wave,FMCW)体制的毫米波雷达系统。通过优化硬件架构... 为了开发一套低成本且高实用性的旋翼无人机合成孔径雷达(Synthetic Aperture Radar,SAR)成像系统,该文以六旋翼无人机为载体,设计了线性调频连续波(Frequency Modulated Continuous Wave,FMCW)体制的毫米波雷达系统。通过优化硬件架构和改进传统SAR信号处理算法,提升了系统性能。进行了多点目标、小型无人船、停车场等场景的实际成像实验,结果表明,该系统在各类场景中的成像分辨率达到0.3 m,证明系统具有较高的成像精度和可靠性。该文为构建低成本SAR成像系统提供了一种有效解决方案。 展开更多
关键词 低成本合成孔径雷达系统 旋翼无人机 线性调频连续波 速度估计算法
在线阅读 下载PDF
Admissible Linear Estimators of Multivariate Regression Coefcient with Respect to an Inequality Constraint under Balanced Loss Function
13
作者 Jie WU Daojiang HE 《Journal of Mathematical Research with Applications》 CSCD 2013年第6期745-752,共8页
In this paper, the admissibility of multivariate linear regression coefficient with respect to an inequality constraint under balanced loss function is investigated. Necessary and sufficient conditions for admissible ... In this paper, the admissibility of multivariate linear regression coefficient with respect to an inequality constraint under balanced loss function is investigated. Necessary and sufficient conditions for admissible homogeneous and inhomogeneous linear estimators are obtained, respectively. 展开更多
关键词 ADMISSIBILITY inequality constraint balanced loss function homogeneous (inhomogeneous) linear estimator.
原文传递
Time Delay Estimation of Target Echo Signal Based on Multi-bright Spot Echoes
14
作者 Ge Yu Fan Du +1 位作者 Xiukun Li Yan Li 《哈尔滨工程大学学报(英文版)》 2026年第1期312-325,共14页
Accurate time delay estimation of target echo signals is a critical component of underwater target localization.In active sonar systems,echo signal processing is vulnerable to the effects of reverberation and noise in... Accurate time delay estimation of target echo signals is a critical component of underwater target localization.In active sonar systems,echo signal processing is vulnerable to the effects of reverberation and noise in the maritime environment.This paper proposes a novel method for estimating target time delay using multi-bright spot echoes,assuming the target’s size and depth are known.Aiming to effectively enhance the extraction of geometric features from the target echoes and mitigate the impact of reverberation and noise,the proposed approach employs the fractional order Fourier transform-frequency sliced wavelet transform to extract multi-bright spot echoes.Using the highlighting model theory and the target size information,an observation matrix is constructed to represent multi-angle incident signals and obtain the theoretical scattered echo signals from different angles.Aiming to accurately estimate the target’s time delay,waveform similarity coefficients and mean square error values between the theoretical return signals and received signals are computed across various incident angles and time delays.Simulation results show that,compared to the conventional matched filter,the proposed algorithm reduces the relative error by 65.9%-91.5%at a signal-to noise ratio of-25 dB,and by 66.7%-88.9%at a signal-to-reverberation ratio of−10 dB.This algorithm provides a new approach for the precise localization of submerged targets in shallow water environments. 展开更多
关键词 Multi-bright spot echoes Time-delay estimation Target echo signal Frequency sliced wavelet transform Fractional order fourier transform
在线阅读 下载PDF
一种改进的M-Estimators基础矩阵鲁棒估计法 被引量:6
15
作者 张洁玉 陈强 +1 位作者 刘复昌 夏德深 《中国图象图形学报》 CSCD 北大核心 2009年第8期1663-1668,共6页
针对原M-Estimators算法完全依赖由线性最小二乘法估计得到的矩阵初始值,精度较低稳定性差的缺点,提出了一种改进的M-Estimators算法。通过考虑匹配点与对应极线的距离,计算求得较原M-Estimators算法更加精确的矩阵初始值,再利用此初始... 针对原M-Estimators算法完全依赖由线性最小二乘法估计得到的矩阵初始值,精度较低稳定性差的缺点,提出了一种改进的M-Estimators算法。通过考虑匹配点与对应极线的距离,计算求得较原M-Estimators算法更加精确的矩阵初始值,再利用此初始值剔除掉原匹配点集中的错误匹配点及坏点,最后运用Torr-M-Estimators法对新的匹配点集进行非线性优化计算,得到了真正的匹配点对,精确恢复了对极几何关系。以大量的模拟数据和真实图像进行了实验,给出了该算法与其他鲁棒性算法的比较结果,实验结果表明,该算法在误匹配以及高斯噪声存在的情况下,提高了基础矩阵的估计精度,并且同时具有很好的鲁棒性。 展开更多
关键词 基础矩阵 鲁棒性 精确初始矩阵 M估计法 最小中值法
在线阅读 下载PDF
不确定性感知的标签噪声矫正算法
16
作者 李英双 贾文玉 +2 位作者 杨莉 曾旺官 董永峰 《郑州大学学报(理学版)》 北大核心 2026年第1期10-18,共9页
标签噪声通过在训练过程中降低对真实类别预测的置信度引入了不确定性问题,为了降低标签噪声的影响,提出了一种不确定性感知的标签噪声矫正算法(ULC)。首先,基于证据理论和主观逻辑理论,从样本的多个视图和标签信息中估计不确定性。其次... 标签噪声通过在训练过程中降低对真实类别预测的置信度引入了不确定性问题,为了降低标签噪声的影响,提出了一种不确定性感知的标签噪声矫正算法(ULC)。首先,基于证据理论和主观逻辑理论,从样本的多个视图和标签信息中估计不确定性。其次,采用双准则样本选择策略将数据集划分为三个子集,并使用联合预测矫正噪声标签。最后,采用不同的正则化策略处理各个子集以优化训练目标。在四个模拟标签噪声数据集和两个真实标签噪声数据集上进行对比实验。与DivideMix算法相比,在包含40%Pairflip类型噪声的CIFAR-10和CIFAR-100数据集上,ULC的分类准确率分别提升了10.58个百分点和15.84个百分点,矫正标签准确率分别达到了95.48%和81.32%。实验结果表明,ULC能够准确估计不确定性,提升矫正标签准确率和模型泛化性能。 展开更多
关键词 深度学习 标签噪声 不确定性估计 样本选择 标签矫正
在线阅读 下载PDF
一种基于稀疏性的RIS辅助信道估计方法
17
作者 刘佳璇 张笑剑 王丹 《计算机应用研究》 北大核心 2026年第1期234-239,共6页
为解决RIS(reconfigurable intelligent surface)辅助通信系统存在的两大核心挑战:高昂的导频资源消耗与异常用户(权重小于等于0.1的用户)导致的信道干扰,改进了基于改进正交匹配追踪算法(OMP)的双稀疏性信道估计方法(D-OMP)。改进算法... 为解决RIS(reconfigurable intelligent surface)辅助通信系统存在的两大核心挑战:高昂的导频资源消耗与异常用户(权重小于等于0.1的用户)导致的信道干扰,改进了基于改进正交匹配追踪算法(OMP)的双稀疏性信道估计方法(D-OMP)。改进算法充分利用级联信道在角度域的双稀疏性,针对D-OMP三阶段估计进行改进:a)联合估计多用户公共行支持,采用中位数能量检测代替均值计算有效抑制异常用户干扰;b)引入多用户加权,优化部分公共列支持;c)引入多用户加权系数自适应剔除异常用户,优化用户特定列支持。仿真结果表明,该算法在减少导频开销的同时,NMSE较传统OMP降低3 dB以上,且在低信噪比和异常用户干扰下保持稳定,显著提升系统鲁棒性。该算法通过双稀疏性联合估计与异常用户抑制机制,为高密度RIS组网提供了低复杂度的信道估计解决方案。 展开更多
关键词 可重构智能反射面 信道估计 级联信道 双稀疏性 正交匹配追踪
在线阅读 下载PDF
Testable Design and BIST Techniques for Systolic Motion Estimators in Transform Domain
18
作者 Shyue-Kung Lu Wei-Yuan Liu 《Journal of Electronic Science and Technology of China》 2009年第4期291-296,共6页
Testable design techniques for systolic motion estimators based on M-testability conditions are proposed in this paper. The whole motion estimator can be viewed as a two-dimensional iterative logic array (ILA) of pr... Testable design techniques for systolic motion estimators based on M-testability conditions are proposed in this paper. The whole motion estimator can be viewed as a two-dimensional iterative logic array (ILA) of processing elements (PEs) and multiplying elements (MULs). The functions of each PE and MUL are modified to be bijective to meet the M-testable conditions. The number of test patterns is 2^w, where w denotes the word length of a PE. The proposed testable design techniques are also suitable for built-in self-test implementation. According to experimental results, our approaches can achieve 99.27 % fault coverage. The area overhead is about 9 %. To verify our approaches, an experimental chip is also implemented. 展开更多
关键词 Built-in self-test design for testability fault coverage motion estimator.
在线阅读 下载PDF
基于密度判别的核均值漂移异常检测
19
作者 吴霜霜 任芮彬 +1 位作者 赵联文 廖群英 《四川师范大学学报(自然科学版)》 2026年第1期114-122,共9页
异常检测是一项重要的数据挖掘任务,旨在找到不符合预期模式的数据.针对原始的均值漂移异常值检测算法(MOD)在非凸、密度不均数据集上的不适用性,提出基于密度判别的核均值漂移异常值检测算法,即KD-kMOD算法.算法首先引入核函数计算均... 异常检测是一项重要的数据挖掘任务,旨在找到不符合预期模式的数据.针对原始的均值漂移异常值检测算法(MOD)在非凸、密度不均数据集上的不适用性,提出基于密度判别的核均值漂移异常值检测算法,即KD-kMOD算法.算法首先引入核函数计算均值漂移向量,使得在有效区域中的不同样本点具有不同的权重,同时降低k值影响;其次,算法在识别出预内值和预异常值的基础上,通过KD树构造数据集,自适应得出基于密度的查询半径r;最后,基于查询半径内数据密度判别异常点.该算法同时考虑全局信息和局部密度,将先前基于距离的异常评判方式转化成为基于密度的判别方式.人工数据集的实验证明,KD-kMOD算法能够适用于非凸、密度不均的数据集,同时在真实数据集上仍能高效稳定地检测出异常值,精确率(Pr)和召回率(Re)均达到85%左右. 展开更多
关键词 核密度估计 均值漂移 异常检测
在线阅读 下载PDF
基于多任务学习的跳频调制方式识别与信噪比估计方法
20
作者 汪有鹏 王昊 曹建银 《现代电子技术》 北大核心 2026年第1期66-72,共7页
针对目前在跳频信号识别的多任务学习中存在跷跷板现象和使用IQ信号训练出的模型泛化能力较差的问题,文中提出一种改进的方法,采用CGC的多任务网络框架结合大卷积核与结构重参数化技术,以提高跳频信号调制识别和信噪比估计的准确性。该... 针对目前在跳频信号识别的多任务学习中存在跷跷板现象和使用IQ信号训练出的模型泛化能力较差的问题,文中提出一种改进的方法,采用CGC的多任务网络框架结合大卷积核与结构重参数化技术,以提高跳频信号调制识别和信噪比估计的准确性。该多任务网络架构采用硬参数共享,将网络通道划分为专家通道和共享通道,并引入了包含大卷积核结构重参数化与残差结构的MobileBlock层。与多任务学习中常用的MMOE结构模型相比,跳频信号调制识别的分类准确率更高,信噪比估计的均方误差更小。实验结果证明了该方法在现代军事通信对抗中的应用潜力,为跳频信号识别和参数估计提供了一个较好的解决方案。 展开更多
关键词 跳频信号 调制识别 信噪比估计 多任务学习 大核卷积 结构重参数化
在线阅读 下载PDF
上一页 1 2 250 下一页 到第
使用帮助 返回顶部