基于卷积神经网络的5G蜂窝网络无线定位方法

5G蜂窝网络发展迅猛，其覆盖面积将逐渐增大，因此使用5G蜂窝网络进行定位是有研究潜力的研究方向。本文提出一种新的深度学习技术来实现高效、高精度和低占用的定位，以代替传统指纹定位过程中繁重的指纹库生成以及距离计算。该方法建立了一个特殊的卷积神经网络，并根据5G天线信号的接收信号强度指示、相位和到达角等特征量，选择合适的输入数据格式构造样本组建训练集，对该卷积神经网络进行训练。训练得到的卷积神经网络可以替代指纹定位中的庞大指纹库，非常有利于直接在5G移动设备端实现定位。虽然卷积神经网络在训练过程中需要大量时间，但在训练完毕后直接进行分类定位的速度非常快，可以保障定位实现的实时性。本文所实现的卷积神经网络权重与偏置所占内存不到0.5 MB，且能够在实际应用环境中以95%的定位准确率以及0.1 m的平均定位精度实现高精度定位。     

基于PBFT区块链技术的电网企业仓储系统

提出了一种基于实用拜占庭容错(PBFT)算法的区块链技术,首先对传统的实用拜占庭容错算法原理进行了阐述,该传统算法包含前期、需求、预准备、准备、确认、答复6个阶段,但传统算法具有实时性差、缺乏惩罚机制、带宽高的缺点。针对出现的这些问题,又对传统算法进行了改进,具体涉及记账节点、共识过程以及视图切换过程。通过测试进一步证明了该改进算法的实用性,并将该算法应用于电网企业中,构建的虚拟仓库实现了联储联备,降低了库存资金的耗费,并且提高了电网企业库存管理的效率。     

干扰条件下的“北斗”信号解扩重构DOA估计

在干扰条件下，卫星导航抗干扰波束形成算法往往需要卫星信号波达方向（Direction-of-Arrival，DOA）的先验信息。但当存在低信噪比信号或主动干扰源时，常规的DOA估计算法性能急剧下降甚至失效。针对此问题，提出了一种被干扰信号压制的低信噪比“北斗”信号的DOA估计算法。该算法首先通过对接收信号进行子空间投影抑制干扰信号，然后对抑制干扰后的信号进行解扩重构处理，最后通过多重信号分类算法完成对“北斗”信号的DOA估计。仿真结果表明，在干扰信号干信比80 dB条件下，“北斗”信号DOA估计误差在5°以内，为下一步进行波束形成计算提供了高精度的入射角信息。     

Deep reinforcement learning-based cooperative interactions among heterogeneous vehicular networks

Most real-world vehicle nodes can be structured into an interconnected network of vehicles. Through structuring these services and vehicle device interactions into multiple types, such internet of vehicles becomes multidimensional heterogeneous overlay networks. The heterogeneousness of the overlays makes it difficult for the overlay networks to coordinate with each other to improve their performance. Therefore, it poses an interesting but critical challenge to the effective analysis of heterogeneous virtual vehicular networks. A variety of virtual vehicular networks can be easily deployed onto the native network by applying the concept of SDN (Software Defined Networking). These virtual networks reflect their heterogeneousness due to their different performance goals, and they compete for the same physical resources of the underlying network, so that a sub-optimal performance of the virtual networks may be achieved. Therefore, we propose a Deep Reinforcement Learning (DRL) approach to make the virtual networks cooperate with each other through the SDN controller. A cooperative solution based on the asymmetric Nash bargaining is proposed for co-existing virtual networks to improve their performance. Moreover, the Markov Chain model and DRL resolution are introduced to leverage the heterogeneous performance goals of virtual networks. The implementation of the approach is introduced, and simulation results confirm the performance improvement of the latency sensitive, loss-rate sensitive and throughput sensitive heterogeneous vehicular networks using our cooperative solution.     

云杉八齿小蠹Ips typographus L.虚拟模型构建方法研究

本文以云杉八齿小蠹Ips typographus Linnaeus为例,经标本选取,观察虫体整体形态,绘制整体结构草图,电子显微镜观察局部、得到局部数字图像,分别建立虫体每一部分的高精度细节。把模型每部分拼装在一起组成整体模型,构建出了云杉八齿小蠹三维虚拟昆虫数字化模型。     

基于信号增强的缓慢泄漏检测方法

目前管道泄漏检测方法可有效检测突发泄漏，对于缓慢泄漏则存在检测灵敏度低、定位不准确等问题。基于此，提出了一种基于信号增强的缓慢泄漏检测方法。通过信号压缩（抽取及移位）克服缓慢泄漏压力信号下降平缓的缺点；根据声波信号具有波形尖锐突出、对突发泄漏敏感的优点，通过建立以压力为输入、虚拟声波为输出的声波信号变送器模型，将压力信号转换为声波信号，克服了泄漏压力信号容易被淹没在管道压力波动及背景噪声中的缺点，实现了缓慢泄漏信号的增强；利用临近插值方法重构虚拟声波信号，基于延时互相关分析实现了缓慢泄漏的准确定位。实验结果表明，该方法具有显著的信号增强效果和定位精度，实现了缓慢泄漏的准确检测。     

Knowledge-Based Virtual Modeling and Simulation of Manufacturing Processes for Industry 4.0

Abstract

Industry 4.0 aims at providing a digital representation of a production landscape, but the challenges in building, maintaining, optimizing, and evolving digital models in inter-organizational production chains have not been identified yet in a systematic manner. In this paper, various Industry 4.0 research and technical challenges are addressed, and their present scenario is discussed. Moreover, in this article, the novel concept of developing experience-based virtual models of engineering entities, process, and the factory is presented. These models of production units, processes, and procedures are accomplished by virtual engineering object (VEO), virtual engineering process (VEP), and virtual engineering factory (VEF), using the knowledge representation technique of Decisional DNA. This blend of the virtual and physical domains permits monitoring of systems and analysis of data to foresee problems before they occur, develop new opportunities, prevent downtime, and even plan for the future by using simulations. Furthermore, the proposed virtual model concept not only has the capability of Query Processing and Data Integration for Industrial Data but also real-time visualization of data stream processing.     

水下图像处理技术研究综述

随着自主式水下机器人的发展，水下探测技术成为新的研究热点。然而，吸收效应和散射效应导致水下获取的图像存在雾化和色彩偏差等缺陷。降质的水下图像在一定程度上降低了水下目标识别的准确性。为了改善水下图像质量，国内外学者对水下图像处理方法进行了深入研究。因水下图像处理方法对提升水下目标识别准确性具有良好的促进作用，故其具有重要的研究与分析价值。介绍了水下成像模型，分析了水下图像视觉质量下降的原理；根据水下物理成像模型将水下图像处理方法分为水下图像增强与水下图像复原，并分别对两类方法的研究现状进行分析与归纳；最后，总结与讨论了各类方法的优缺点，并展望了未来的发展方向。     

Reality-based interaction affecting mental workload in virtual reality mental arithmetic training

ABSTRACT

The concept of digital game-based learning (DGBL) evolves rapidly together with technological enhancements of virtual reality (VR) and smart phones. However, the mental workload (MWL) that VR-training applications demand and motivational qualities originating from user experience (UX) should be identified in order to create effective and enjoyable training/learning challenges that fit with individual users’ capabilities. This study examined the effects of reality-based interaction (RBI) and VR on measures of student motivation and MWL, in a mental arithmetic game for secondary school pupils. In a randomised controlled trial with sixty school children, a mental arithmetic game was tested with three different interaction and two different presentation methods – VR RBI, VR head-mounted-display tapping and tablet flick-gesture. Results found a significant effect of RBI on MWL but no differences in enjoyment of training were found between VR-experience and tablet training-experience. In fact, adding the gaming-context to the mental arithmetic task created an enjoyable, motivating experience regardless of presentation or interaction-style.     

应用膨胀算法提取目标的直接定位算法

为了解决被动雷达系统中的多发射源定位问题，提出了一种基于多重信号分类(MUSIC)算法和图像膨胀(IE)算法的直接定位方法。该方法结合了谱分析中的MUSIC思想，通过对接收量测协方差矩阵进行特征分析求解目标的位置。首先，在目标个数未知的前提下，利用Akaike信息准则(AIC)来确定模型阶数；然后，推导了基于MUSIC的定位代价函数；之后，利用图像膨胀算法处理得到的代价函数平面；最后，膨胀处理后的输出为目标个数及目标位置的估计值。提出的算法有效地解决了目标检测及提取的问题，能够确定多个目标的位置坐标，为后续的定位性能分析提供可能性，也保证了算法的完整性。进一步地分析了多个临近目标情况下影响目标提取性能的主要因素。