无双线性对的无证书安全签密方案

已有的无双线性对的无证书签密方案,大都存在着机密性漏洞和签名选择性伪造问题。针对此问题,基于有限乘法交换群上的离散对数难题,提出一种无双线性对的无证书签密方案,并进行了安全分析和证明。指出采用随机预言模型的证明方法可能忽略不解决DL难题时对签密方案的攻击,不能证明破解签密方案的难度与解决DL难题相等价。通过对参数的数学分析,证明了破解本签密方案的必要条件是解决DL难题,不存在不解决DL难题攻击本签密方案的多项式时间算法。与已知所有方案相比,所提方案安全性更高。     

自适应斥力系数的无人机路径规划

使用人工势场法进行无人机路径规划时,往往存在目标不可达、运动轨迹迂回反复和路径长度过长等问题.传统的人工势场法不能根据环境具体信息对斥力系数进行调整,而现有的改进方法不能在自适应调整斥力系数的同时兼顾规划效果和规划时长.针对以上问题,提出了一种基于深度学习的无人机自适应斥力系数路径规划方法.首先通过融合遗传算法与人工势场法找出在特定环境下最合适的斥力系数样本集,其次利用该样本集训练残差神经网络,最后通过残差神经网络计算适应环境的斥力系数,进而使用人工势场法进行路径规划.仿真实验表明,该方法在一定程度上解决了人工势场法规划中目标不可达、运动轨迹迂回反复和路径长度过长等问题,规划效果和规划时长方面均有优异表现,能很好地满足无人机路径规划中对当前环境的自适应要求和快速规划的要求.     

Bi-directional LSTM based channel estimation in 5G massive MIMO OFDM systems over TDL-C model with Rayleigh fading distribution

In this work, a deep learning (DL)-based massive multiple-input multiple-output (mMIMO) orthogonal frequency division multiplexing (OFDM) system is investigated over the tapped delay line type C (TDL-C) model with a Rayleigh fading distribution at frequencies ranging from 0.5 to 100 GHz. The proposed bi-directional long short-term memory (Bi-LSTM) channel state information (CSI) estimator uses online learning during training and offline learning during the practical implementation phase. The design of the estimator takes into account situations in which prior knowledge of channel statistics is limited and targets excellent performance, even with limited pilot symbols (PS). Three separate loss functions (mean square logarithmic error [MSLE], Huber, and Kullback–Leibler Distance [KLD]) are assessed in three classification layers. The symbol error rate (SER) and outage probability performance of the proposed estimator are evaluated using a number of optimization techniques, such as stochastic gradient descent (SGD), momentum, and the adaptive gradient (AdaGrad) algorithm. The Bi-LSTM-based CSI estimator is trained considering a specific number of PS. It can be readily seen that by incorporating a cyclic prefix (CP), the system becomes more resilient to channel impairments, resulting in a lower SER. Simulations show that the SGD optimization approach and Huber loss function-trained Bi-LSTM-based CSI estimator have the lowest SER and very high estimation accuracy. By using deep neural networks (DNNs), the Bi-LSTM method for CSI estimation achieves a superior channel capacity (in bps/Hz) at 10 dB than long short-term memory (LSTM) and other conventional CSI estimators, such as minimum mean square error (MMSE) and least squares (LS). The simulation results validate the analytical results in the study.     

基于对抗深度学习的无人机航拍违建场地识别

对违建场地的检测方法主要是通过人工对无人机航拍视频进行检查,存在检测精度低、识别性能差、工作效率低的问题。提出一种结合空间变换网络与Fast RCNN的生成对抗网络ASTN-Fast RCNN,通过深度学习与无人机航拍视频相结合自动识别检测处在建设初期的违建场地。将空间变换网络作为生成器,生成Fast RCNN目标检测器难以识别的旋转形变样本,并通过目标检测器与生成器的对抗式训练,提高检测器的鲁棒性。实验结果表明,该方法能够有效提高对无人机航拍违建场地的识别性能。     

一种基于YOLOv4-TIA的林业害虫实时检测方法

针对现有基于深度学习的林业昆虫图像检测方法存在检测精度低和检测速度慢的问题,提出一种结合改进PANet结构与三分支注意力机制的目标检测方法YOLOv4-TIA。通过对样本数量较少的昆虫类别进行数据增强,实现样本均衡分布。利用三分支注意力机制改进YOLOv4中的CSPDarkNet53骨干网络,同时通过旋转操作和残差变换建立维度间的依存关系,以提高有效的特征通道权重,在PANet结构上增加将跳跃连接与跨尺度连接相结合的特征融合方式,从而获取更丰富的语义信息和位置信息。在此基础上,采用Focal loss函数优化分类损失,解决正负样本不均衡的问题。实验结果表明,该方法的精确率和召回率分别达到85.9%和91.2%,相比SSD、Faster R-CNN、YOLOv4方法,其在保证检测速度的同时,能够有效提高检测精度,且实现对林业害虫的实时精确监测。     

A formally specified ontology management API as a registry for ubiquitous computing systems

Recently, several standards have emerged for ontology markup languages that can be used to formalize all kinds of knowledge. However, there are no widely accepted standards yet that define APIs to manage ontological data. Processing ontological information still suffers from the heterogeneity imposed by the plethora of available ontology management systems. Moreover, ubiquitous computing environments usually comprise software components written in a variety of different programming languages, which makes it particularly difficult to establish a common ontology management API with programming language agnostic semantics. We implemented an ontological Knowledge Base Server, which can expose the functionality of arbitrary off-the-shelf ontology management systems via a formally specified and well defined API. A case study was carried out in order to demonstrate the feasibility of our approach to use a formally specified ontology management API to implement a registry for ubiquitous computing systems.     

支持OAI-PMH的元数据互操作体系结构设计与实现

随着全球数字图书馆的迅猛发展,互操作成为亟待解决的重要问题。OAI-PMH协议的制定为在元数据级别上设计实现一个统一、高效的互操作框架创造了条件。文章基于北京大学数字图书馆项目,详尽阐述了一个典型的支持OAI-PMH互操作体系结构的完整的设计与实现;不仅包括在理论上逻辑清晰的设计方案,在实现上也采用了比较成熟又兼具理论和应用前景的技术,如Java、DOM、XML、Oracle9i等。该文的工作为将来联合式数字图书馆这一发展方向做出了很好的尝试。     

Performance of supercritical carbon dioxide sprays as coolants and lubricants in representative metalworking operations

This paper investigates the cooling and lubrication properties of supercritical carbon dioxide (scCO₂) sprays as potential substitutes for aqueous emulsions and straight oils used in the metalworking industry today. Sprays of rapidly expanding scCO₂ act to cool and lubricate machining and forming processes by delivering a mixture of dry ice and lubricant deep into the cutting/forming zone. In this work, experiments with turning, milling, drilling, thread cutting, and thread forming were performed with scCO₂ and other metalworking fluids (MWFs) to evaluate their relative performance with respect to tool wear and machining torque. Observations reveal that scCO₂–MWFs are more effective in removing heat from the tool-workpiece interface than conventionally delivered (flood) aqueous MWFs as well as other gas-based MWF sprays. In addition, scCO₂–MWFs delivered in lubricant-expanded phase, where scCO₂ is used to increase volume of lubricant in the spray field, are shown to provide better lubricity than straight oils and oil-in-air minimum quantity lubrication (MQL) sprays. As a result, scCO₂–MWFs can reduce tool wear and improve machining productivity in a wide range of manufacturing operations leading to appreciable improvements in the economics of manufacturing. Also given that CO₂ is a recovered waste gas that is non-toxic, scCO₂–MWFs can improve the environmental and worker health performance of manufacturing operations.     

Efficient loading and entrapment of tamoxifen in human serum albumin based nanoparticulate delivery system by a modified desolvation technique

Herein, the poorly water-soluble drug, Tamoxifen (Tmx), was loaded in the amphipathic matrix of human serum albumin (HSA) nanoparticles by a modified desolvation method. In order to enhance the drug loading (DL) and drug entrapment efficiency (DEE) (<2% and 10%, respectively), ultrasonication of Tmx-HSA mixture was performed prior to desolvation process. Tmx loading and entrapment efficiency were optimized by employment of the response surface methodology (RSM)-central composite design (CCD) of experiments. Under the optimum conditions of 1.59 mg Tmx/ml concentration, 7.76 pH and 5 h incubation of HSA-Tmx, the DL of 6.7% and DEE of 74% are achievable. Particles with the average size of 195 nm, zeta potential of −21 mV and polydispersity index of 0.09 were produced under these conditions. A more sustained Tmx release behavior was observed from polyethylene glycol (PEG) conjugated nanoparticles in comparison to the non-PEGylated ones. The short-term stability investigation showed no alteration in physicochemical properties of nanoparticles at 4 and 37 °C, but small increase in nanoparticles size was observed after three months of storage at room temperature. This is the first report for efficient production of a Tmx delivery system based on HSA nanoparticles.     

基于DL联合信息融合技术的刮板转载机故障诊断

采用深度学习(DL)算法和信息融合技术对刮板转载机进行故障诊断。对刮板转载机故障类型进行分类,建立刮板转载机故障诊断模型。为提高信息融合技术的准确率与效率,将一种典型的DL算法——深度置信网络(DBN)应用于数据层信息融合,提出基于DL联合信息融合技术的故障诊断流程。对比MATLAB仿真结果与真实情况,基于DL联合信息融合技术故障诊断结果的准确率比信息融合技术的高。