In heavy-duty long-distance transmission scenarios,steel wire ropes are widely used due to their unique advantages,and their safety is very important,which has also led to the rapid development of non-destructive test...In heavy-duty long-distance transmission scenarios,steel wire ropes are widely used due to their unique advantages,and their safety is very important,which has also led to the rapid development of non-destructive testing technology for steel wire ropes.The non-destructive testing technology for steel wire ropes is influenced by various factors such as its own structure and external working environment,and the testing process is relatively complex.Multiple testing methods and related types of sensors have also emerged.The electromagnetic detection method is currently the most effective method,but it also has its limitations in development and has not yet fully achieved the expected detection goals.In order to completely replace manual inspection work with the development of non-destructive testing technology for steel wire ropes,more in-depth research and long-term accumulation are still needed.展开更多
The measurement of the rolling angle of the projectile is one of the key technologies for the terminal correction projectile.To improve the resolution accuracy of the rolling angle in the laser seeker weapon system, t...The measurement of the rolling angle of the projectile is one of the key technologies for the terminal correction projectile.To improve the resolution accuracy of the rolling angle in the laser seeker weapon system, the imaging model of the detector, calculation model of the position and the signal-to-noise ratio(SNR) model of the circuit are built to derive both the correlation between the resolution error of the rolling angle and the spot position, and the relation between the position resolution error and the SNR. Then the influence of each parameter on the SNR is analyzed at large,and the parameters of the circuit are determined. Meanwhile, the SNR and noise voltage of the circuit are calculated according to the SNR model and the decay model of the laser energy. Finally,the actual photoelectric detection circuit is built, whose SNR is measured to be up to 53 d B. It can fully meet the requirement of0.5° for the resolution error of the rolling angle, thereby realizing the analysis of critical technology for photoelectric detection of laser seeker signals.展开更多
音符起始点是音乐特征信息中最基础的特征,也是高级音乐分析任务(如基频估计、节奏分析、节拍跟踪等)的前提.然而,目前大多数音符起始点检测算法的研究对象多以西洋乐(如钢琴、小提琴等)或人声为主,针对中国民乐的研究极为匮乏,其主要...音符起始点是音乐特征信息中最基础的特征,也是高级音乐分析任务(如基频估计、节奏分析、节拍跟踪等)的前提.然而,目前大多数音符起始点检测算法的研究对象多以西洋乐(如钢琴、小提琴等)或人声为主,针对中国民乐的研究极为匮乏,其主要原因是缺少高质量的带标注的数据集.为此,本文采集了古筝的部分考级曲目的音频,并在专家指导下对音符起始点加以标注,率先构建了一个带标签的包含3529条记录的古筝音符起始点数据集.此外,由于传统的神经网络特征预处理方法无法凸显古筝音符的起始点特征,本文提出了一种新型的特征平滑处理方法ELES(Extremum filtering based envelope smoothing),对训练数据的均值和标准差进行平滑处理,再经过标准化方法强化起始点特征,随后送入卷积神经网络进行训练.实验证明本文方法的准确性优于传统的基于音频信号处理的方法,并能够有效地检测出古筝的特殊技法和连续十六分音符的起始点.展开更多
物联网(Internet of Things,IoT)无处不在的、连接的、智能的节点可以自主交互,同时提供各种服务,其分布广、开放性强、处理能力相对较高的特点也使它们成为网络攻击的理想目标。此外,由于许多物联网节点在收集和处理私人信息,使得它们...物联网(Internet of Things,IoT)无处不在的、连接的、智能的节点可以自主交互,同时提供各种服务,其分布广、开放性强、处理能力相对较高的特点也使它们成为网络攻击的理想目标。此外,由于许多物联网节点在收集和处理私人信息,使得它们成为恶意行为者的重要数据来源。因此,安全性,特别是检测受感染节点的能力以及收集和保存攻击或恶意活动的证据,成为物联网成功部署的优先事项。基于此,介绍物联网领域现有的安全和取证方面的挑战及机遇。展开更多
文摘针对去蜂窝大规模多输入多输出(Cell-Free massive Multiple Input Multiple Output,CF-mMIMO)系统多用户场景下导频污染抑制所涉及的优化目标维度高的问题,提出基于K-Means分簇和用户优先级的导频分配方案。该方法基于指数态聚类误差平方和(Sum of the Squared Error,SSE)确定最佳K值,再基于大尺度衰落进行簇心初始化,以此改进K-Means方法实现用户分簇;然后根据分簇结果先基于类高斯距离对簇交界处高优先级用户分配导频,再对每个簇内低优先级用户依据最大误差最小化独立分配导频。这种方法可以显著降低导频污染抑制问题维度,改善导频分配机制。仿真结果表明,所提方法可以有效抑制多用户场景下的导频污染,提高吞吐量,并降低干扰抑制问题的维度。
文摘In heavy-duty long-distance transmission scenarios,steel wire ropes are widely used due to their unique advantages,and their safety is very important,which has also led to the rapid development of non-destructive testing technology for steel wire ropes.The non-destructive testing technology for steel wire ropes is influenced by various factors such as its own structure and external working environment,and the testing process is relatively complex.Multiple testing methods and related types of sensors have also emerged.The electromagnetic detection method is currently the most effective method,but it also has its limitations in development and has not yet fully achieved the expected detection goals.In order to completely replace manual inspection work with the development of non-destructive testing technology for steel wire ropes,more in-depth research and long-term accumulation are still needed.
基金supported by the National Natural Science Foundation of China(61201391)
文摘The measurement of the rolling angle of the projectile is one of the key technologies for the terminal correction projectile.To improve the resolution accuracy of the rolling angle in the laser seeker weapon system, the imaging model of the detector, calculation model of the position and the signal-to-noise ratio(SNR) model of the circuit are built to derive both the correlation between the resolution error of the rolling angle and the spot position, and the relation between the position resolution error and the SNR. Then the influence of each parameter on the SNR is analyzed at large,and the parameters of the circuit are determined. Meanwhile, the SNR and noise voltage of the circuit are calculated according to the SNR model and the decay model of the laser energy. Finally,the actual photoelectric detection circuit is built, whose SNR is measured to be up to 53 d B. It can fully meet the requirement of0.5° for the resolution error of the rolling angle, thereby realizing the analysis of critical technology for photoelectric detection of laser seeker signals.
文摘音符起始点是音乐特征信息中最基础的特征,也是高级音乐分析任务(如基频估计、节奏分析、节拍跟踪等)的前提.然而,目前大多数音符起始点检测算法的研究对象多以西洋乐(如钢琴、小提琴等)或人声为主,针对中国民乐的研究极为匮乏,其主要原因是缺少高质量的带标注的数据集.为此,本文采集了古筝的部分考级曲目的音频,并在专家指导下对音符起始点加以标注,率先构建了一个带标签的包含3529条记录的古筝音符起始点数据集.此外,由于传统的神经网络特征预处理方法无法凸显古筝音符的起始点特征,本文提出了一种新型的特征平滑处理方法ELES(Extremum filtering based envelope smoothing),对训练数据的均值和标准差进行平滑处理,再经过标准化方法强化起始点特征,随后送入卷积神经网络进行训练.实验证明本文方法的准确性优于传统的基于音频信号处理的方法,并能够有效地检测出古筝的特殊技法和连续十六分音符的起始点.
文摘物联网(Internet of Things,IoT)无处不在的、连接的、智能的节点可以自主交互,同时提供各种服务,其分布广、开放性强、处理能力相对较高的特点也使它们成为网络攻击的理想目标。此外,由于许多物联网节点在收集和处理私人信息,使得它们成为恶意行为者的重要数据来源。因此,安全性,特别是检测受感染节点的能力以及收集和保存攻击或恶意活动的证据,成为物联网成功部署的优先事项。基于此,介绍物联网领域现有的安全和取证方面的挑战及机遇。
