我国天地一体化航天互联网构想

地面互联网技术的快速发展以及航天任务的复杂性不断提高，产生了将互联网扩展到空间、建立天地一体化航天互联网的强烈需求，并提供了实现的技术可能性。探讨了我国航天任务中存在的不同卫星天地协议不统一、航天任务数据不能共享、资源综合利用率低等问题，分析了对未来任务的需求，阐述了建设我国天地一体化航天互联网的重要意义；介绍了国外航天互联网发展状况；提出了我国天地一体化航天互联网的总体目标、组成、网络体系结构及网络协议的初步构想；分析了需要突破的关键技术；最后提出了发展步骤建议。     

Research on Sensor Network Coverage Enhancement Based on Non-Cooperative Games

Coverage is an important issue for resources rational allocation, cognitive tasks completion in sensor networks. The mobility, communicability and learning ability of smart sensors have received much attention in the past decade. Based on the deep study of game theory, a mobile sensor non-cooperative game model is established for the sensor network deployment and a local information-based topology control (LITC) algorithm for coverage enhancement is proposed. We both consider revenue of the monitoring events and neighboring sensors to avoid nodes aggregation when formulating the utility function. We then prove that the non-cooperative game is an exact potential game in which Nash Equilibrium exists. The proposed algorithm focuses on the local information of the neighboring sensors and decides sensors’ next action based on the actions of the other sensors, which maximizes its own utility function. We finally evaluate the performance of the proposed method through simulations. Simulation results demonstrate that the proposed algorithm can enlarge the coverage of the entire monitoring area while achieving effective coverage of the events.     

一种基于HLA/RTI的卫星组网仿真系统

设计了一种基于HLA/RTI的卫星组网仿真系统,在该系统中卫星节点、星间链路、卫星轨道和通信节点路由被划分为功能独立的联邦成员.这种结构具有良好的可扩展性,可适用于不同的卫星星座结构,同时可以适应不同的路由策略和各类空间通信链路模型.最后通过统计卫星组网网管报文响应时间验证了该仿真系统的可行性.卫星组网仿真系统为卫星组网的其他研究提供了仿真验证平台.     

卫星移动网络总体概念设计仿真

卫星移动网络是以卫星星座为基本物理架构，通过星际的通信链路形成覆盖全球的天基通信网络。它对社会发展、科技进步以及军队的信息化有着重大的推动作用。由于天网的移动性、通信的偶断性、接入的复杂性以及使用的综合性，都为研制提出了重大的挑战。有必要通过总体概念设计仿真，来进一步验证总体需求、总体方案、关键技术指标和研制的约束条件，形成科学的可行性报告。介绍了卫星移动网络的总体概念，设计了针对天网仿真的平台，并用实例获得卫星移动网络总体概念设计的仿真数据。     

基于特征金字塔的多尺度特征融合网络

特征金字塔网络（FPN）是CNN网络对图像信息进行表达输出的一种有效方法,在目标检测网络中广泛应用．然而,FPN没有充分地将浅层的细节信息传递到深层的语义特征,存在特征融合不足的缺陷,因而只能依靠深层语义信息来进行预测,从而忽略了网络低层细节信息,对各种视觉学习的效果造成了一定的影响．针对FPN存在的以上问题,本文提出基于特征金字塔的多尺度特征融合网络模型,在FPN主干网络的基础上,设计了混合特征金字塔和金字塔融合模块,并结合注意力机制,对特征金字塔进行了多尺度的深度融合．本文在PASCAL VOC2012和MS COCO2014数据集上,以Faster R-CNN作为基础检测器进行实验,验证了MFPN对特征融合的有效性．     

Dynamic Resource Pricing and Allocation in Multilayer Satellite Network

The goal of delivering high-quality service has spurred research of 6G satellite communication networks. The limited resource-allocation problem has been addressed by next-generation satellite communication networks, especially multilayer networks with multiple low-Earth-orbit (LEO) and non-low-Earth-orbit (NLEO) satellites. In this study, the resource-allocation problem of a multilayer satellite network consisting of one NLEO and multiple LEO satellites is solved. The NLEO satellite is the authorized user of spectrum resources and the LEO satellites are unauthorized users. The resource allocation and dynamic pricing problems are combined, and a dynamic game-based resource pricing and allocation model is proposed to maximize the market advantage of LEO satellites and reduce interference between LEO and NLEO satellites. In the proposed model, the resource price is formulated as the dynamic state of the LEO satellites, using the resource allocation strategy as the control variable. Based on the proposed dynamic game model, an open-loop Nash equilibrium is analyzed, and an algorithm is proposed for the resource pricing and allocation problem. Numerical simulations validate the model and algorithm.     

我国航天测控系统的现状与发展

随着我国航天任务的日益复杂化、多样化，对测控的需求也向“高、精、尖”方向发展，航天测控系统再一次面临新的机遇与挑战。在综合分析国内外航天测控发展历程与现状的基础上，总结了航天测控的发展特点与发展趋势，提出了符合我国国情的“优化地基测控网、建设和发展天基测控网、构建深空探测网，循序渐进构建天地空一体化测控通信系统”的整体发展思路，并对我国地基、天基、深空探测测控通信系统发展及重点技术进行了分析探讨。     

IoT-based networking for humanoid robots

Abstract

In this paper, a new network communication approach, named real-time network (RTNET), is designed and implemented for humanoid robots. The proposed five network objects – alarm, condition, message, mail, and file are used to represent the task and priority of the communication data. Compared to the existing protocols, the network scheduling mechanism of RTNET arranges, more efficiently, the priority and flow control of the five network communication objects to meet real-time requirements for the limited bandwidth of the local area network. RTNET can be further integrated with other protocols, such as EtherCAT or controller area networks (CAN Bus) for local control systems, e.g. robot arms, to improve the communication mechanism. The RTNET can also be used over Ethernet to connect each subsystem and to exchange information among those systems. Also, an Internet of things (IoT) network structure based on RTNET is proposed in this paper. The information of each subsystems is collected through RTNET and users can access all components in the IoT network. In this paper, the concept of RTNET is presented and RTNET has been implemented on a National Taiwan University (NTU) humanoid robot control system with CAN Bus.     

面向电力物联网URLLC业务的智能网络切片管理方法

基于5G通信技术的电力物联网正在如火如荼地建设,随之产生的是用电信息采集、输变电状态监测以及精准负荷控制等新型电力物联网业务。为了满足这些业务对5G网络的超低时延和超高可靠性的需求,提出一种面向电力物联网URLLC(ultra reliable low latency communication)业务的智能网络切片管理方法。该方法综合运用5G切片和移动边缘计算(mobile edge computing,MEC)技术,建立电力业务传输和计算的时延、能耗以及可靠性模型,并通过DQN(deep Q network)算法对切片资源进行优化。仿真实验表明,所提出的智能网络切片管理方法的可靠性高于98%,且优于经典的基于坐标块下降方法和资源平均分配方法。     

改进的蚁群算法及其在卫星网络路由计算中的应用

介绍了蚁群算法的原理，然后对现有蚁群算法进行了一些改进，使它能够快速地收敛以满足高速变化的卫星网络拓扑结构．采用改进的虚拟拓扑策略解决了卫星网络拓扑高速变换的问题．将改进的蚁群算法应用于其上，并给出了相应的性能评估．所提出的改进的虚拟拓扑策略，能够大大减少一个系统周期内卫星网的时间片个数．应用于此基础上的改进的蚁群算法也体现了较好的性能．     

Ground Nephogram Recognition Algorithm Based on Selective Neural Network Ensemble

In view of the low accuracy of traditional ground nephogram recognition model, the authors put forward a k-means algorithm-acquired neural network ensemble method, which takes BP neural network ensemble model as the basis, uses k-means algorithm to choose the individual neural networks with partial diversities for integration, and builds the cloud form classification model. Through simulation experiments on ground nephogram samples, the results show that the algorithm proposed in the article can effectively improve the Classification accuracy of ground nephogram recognition in comparison with applying single BP neural network and traditional BP AdaBoost ensemble algorithm on classification of ground nephogram.     

网络媒体环境下的视觉传达设计探究

目的探索网络媒体环境下视觉传达设计的构成要素与发展策略。方法对比传统的视觉传达设计与网络媒体环境下视觉传达设计,探索两者的诸多不同之处,引出网络媒体环境下视觉传达设计所展现出的强大生命力,然后借助网络媒体的特征与其在视觉传达设计中表现出的自由、交互和实效等方面的突出优势,就其在页面布局、视觉流程、图像处理、色彩搭配等细节上的具体要求展开探索,并结合网络媒体环境下的页面设计和广告设计进行深入论述,探索这一大背景下视觉传达设计的未来发展。结论网络时代的发展促使网络媒体成为视觉传达的主要载体,这不仅带来了视觉传达设计的变化,丰富了设计风格、设计思维方式,还凭借神奇的个性风貌影响着人们的生活,满足了信息时代背景下人们对信息的现实需求。     

基于深度神经网络的素色布匹瑕疵检测算法研究

目的为了改进当前布匹检测算法覆盖瑕疵种类不全、瑕疵检测准确率低和定位精度差的问题,提出一种端到端的素色布匹瑕疵检测的实用算法。方法首先通过图像增强扩充样本数量,使用以Resnet50为主干的Cascade-RCNN网络,通过加入可变形卷积、特征融合网络,增加锚框数目的方法实现素色布匹瑕疵检测。结果通过实验对比表明,该算法可实现检测20种布匹瑕疵,检测是否为瑕疵布匹的准确率为97%,瑕疵定位的平均检测精度为65%,每张样本平均时间为80 ms。结论该算法有效提升了布匹瑕疵检测的准确率和精度,检测瑕疵类别更全面,并且可以获取缺陷位置和类别,能够满足工业上的生产需求。     

MINE: A Method of Multi-Interaction Heterogeneous Information Network Embedding

Interactivity is the most significant feature of network data, especially in social networks. Existing network embedding methods have achieved remarkable results in learning network structure and node attributes, but do not pay attention to the multiinteraction between nodes, which limits the extraction and mining of potential deep interactions between nodes. To tackle the problem, we propose a method called MultiInteraction heterogeneous information Network Embedding (MINE). Firstly, we introduced the multi-interactions heterogeneous information network and extracted complex heterogeneous relation sequences by the multi-interaction extraction algorithm. Secondly, we use a well-designed multi-relationship network fusion model based on the attention mechanism to fuse multiple interactional relationships. Finally, applying a multitasking model makes the learned vector contain richer semantic relationships. A large number of practical experiments prove that our proposed method outperforms existing methods on multiple data sets.     

基于粒子群算法的水声通信网络能量优化方法

水声通信网络节点功耗是影响网络节点寿命的重要因素之一。针对水声通信网络的能量优化问题,基于网络节点发射功率与传输距离的非线性关系,结合改进的粒子群算法建立了一种能量优化方法。该方法改进了网络模型,在网络运行中根据存活节点数量和节点剩余能量的变化情况,自适应动态优化每个节点的信息传输路径。仿真结果表明,所提优化方法能有效降低网络节点总功耗,延缓首个节点的死亡,减缓网络中节点的死亡速率,也即减缓了网络有效覆盖面积随着网络运行而减小的速率。     

基于声信号和一维卷积神经网络的电机故障诊断研究

针对电机故障诊断问题,设计一种新型的一维卷积神经网络结构(1D-CNN),提出一种基于声信号和1DCNN的电机故障诊断方法。为了验证1D-CNN算法在电机故障识别领域的有效性,以一组空调故障电机作为实验对象,搭建电机故障诊断平台,对4种状态的空调电机进行声信号采集实验,制作电机故障声信号数据集,并运用1DCNN算法对数据集进行分类,计算出基于该算法的电机故障识别准确率。实验结果表明,1D-CNN算法作为一种新型结构深度学习算法,能够对电机故障声信号进行很好分类,分类准确率高于FFT-BP、SVM、FFT-SAE等算法。为了探究1D-CNN算法内在机制,还对1D-CNN算法性能进行t-SNE可视化分析。     

Novel DDoS Feature Representation Model Combining Deep Belief Network and Canonical Correlation Analysis

Distributed denial of service (DDoS) attacks launch more and more frequently and are more destructive. Feature representation as an important part of DDoS defense technology directly affects the efficiency of defense. Most DDoS feature extraction methods cannot fully utilize the information of the original data, resulting in the extracted features losing useful features. In this paper, a DDoS feature representation method based on deep belief network (DBN) is proposed. We quantify the original data by the size of the network flows, the distribution of IP addresses and ports, and the diversity of packet sizes of different protocols and train the DBN in an unsupervised manner by these quantified values. Two feedforward neural networks (FFNN) are initialized by the trained deep belief network, and one of the feedforward neural networks continues to be trained in a supervised manner. The canonical correlation analysis (CCA) method is used to fuse the features extracted by two feedforward neural networks per layer. Experiments show that compared with other methods, the proposed method can extract better features.     

空间科学地面系统标准体系建设研究

“十二五”期间是我国空间科学高速发展的阶段,随着空间科学卫星的增多以及卫星寿命的延长,迫切需要建立面向空间科学任务的通用地面应用系统。本文探讨了空间科学卫星任务的特点,以及空间科学卫星地面系统标准化研究的意义。探索了空间科学卫星地面段的总体架构以及各部分的功能和标准要求,从横向业务类别和纵向业务流程两个方面,研究了地面系统的技术标准体系,为建立统一的面向多任务的空间科学卫星地面系统平台奠定技术基础。     

System Performance of Wireless Sensor Network Using LoRa–Zigbee Hybrid Communication

Wireless sensor network (WSN) is considered as the fastest growing technology pattern in recent years because of its applicability in varied domains. Many sensor nodes with different sensing functionalities are deployed in the monitoring area to collect suitable data and transmit it to the gateway. Ensuring communications in heterogeneous WSNs, is a critical issue that needs to be studied. In this research paper, we study the system performance of a heterogeneous WSN using LoRa–Zigbee hybrid communication. Specifically, two Zigbee sensor clusters and two LoRa sensor clusters are used and combined with two Zigbee-to-LoRa converters to communicate in a network managed by a LoRa gateway. The overall system integrates many different sensors in terms of types, communication protocols, and accuracy, which can be used in many applications in realistic environments such as on land, under water, or in the air. In addition to this, a synchronous management software on ThingSpeak Web server and Blynk app is designed. In the proposed system, the token ring protocol in Zigbee network and polling mechanism in LoRa network is used. The system can operate with a packet loss rate of less than 0.5% when the communication range of the Zigbee network is 630 m, and the communication range of the LoRa network is 3.7 km. On the basis of the digital results collected on the management software, this study proves tremendous improvements in the system performance.     

Adaptive optical transmitter and receiver for space communication through thin clouds

Optical space communication from satellite to ground or air to air consists of clouds as part of communication channels. Propagation of optical pulses through clouds causes widening and deformation in the time domain and attenuation of the pulse radiant power. These effects decrease the received signal and limit the information bandwidth of the communication system. Having dealt with the other effects previously, here we concentrate on pulse broadening in the time domain. We derive a mathematical model of an adaptive optical communication system with a multiscattering channel (atmospheric cloud). We use knowledge about the impulse response function of the cloud to adapt the communication parameters to the transfer function of the cloud. The communication system includes a receiver and a transmitter. We adapted the transmitter to atmospheric conditions by changing the bit error rate. One can adapt the receiver to the atmospheric condition by changing the parameters of the detector and the filter. An example for a practical communication system between a low Earth orbit satellite and a ground station cover by cloud is given. Comparison and analysis of an adaptive and semiadaptive system with cloud channels are presented. Our conclusion is that in some cases only by such adaptive methods is optical communication possible.