为进一步提高用户室外指纹定位的预测精度,提出了一种基于LightGBM(轻度提升机)算法和LSTM(长短期记忆神经网络)相结合的模型,通过分析比较基站收集到的用户信息,分别利用LightGBM与LSTM算法进行用户位置的预测,对两种方法预测的结果用误差倒数法减小误差同时对权重进行更新,得到结果作为新的预测值,在用户真实数据集上进行测试.测试结果表明,在对基站收集的用户信息进行指纹定位时,与其它模型相比,文章提出的模型在运行时间和预测精度方面均表现出良好的性能.

针对汽车CAN总线的安全缺陷,通过对计数器、随机数、密钥矩阵、消息认证码与扩展帧的综合利用,提出一种车载CAN总线信息安全机制,使CAN总线具备节点认证与通信加密的能力,从而保证信息的机密性、完整性与真实性.搭建实验平台对所提出的认证与加密机制进行测评与分析,结果表明,在128 MHz的总线频率下,该机制在表现出较快的节点认证速度的同时,安全CAN总线单帧报文的收发可在2 ms之内完成,适应当下汽车的高速发展对CAN总线通信速度的要求.

车联网的迅速发展加速了车辆的智能化进程,作为车内网与车外网通信的源头,车载终端T-BOX在车联网通信中扮演了举足轻重的角色.目前市面上主流的车载T-BOX系统可以通过无线传输模块将行车位置信息上传到远程通信服务提供商(telematics service provider,TSP)服务平台,TSP服务平台再通过其定位功能实时显示车辆地理位置,网联车也可以通过T-BOX实现对其他移动终端的通信.T-BOX在给用户带来便利的同时,也迎来了极大的信息安全风险.为了防止T-BOX遭受到信息安全问题,文章采用了适用于T-BOX与移动终端之间的认证体制及数据加密算法.不同于传统的公钥基础设施(public key infrastructure,PKI),文章采用组合公钥(combined public key,CPK)认证体制,可以通过计算出用户的公钥和私钥,解决超大规模网络中密钥管理难题.

为了解决视频信息遭到非法查看、篡改,平台遭到非法访问等方面的问题,提出了一种基于加密和身份认证的安全的视频监控系统.该系统通过对传统的视频监控信息进行安全与防护设计,如:接入密钥管理系统、数字证书系统、安全视频监控管理平台等,实现视频数据从采集点开始进行加密,保障视频数据在传输、存储过程中的安全性.同时,认证过程保证了数据是合法用户进行获取,确保整个视频监控系统数据不被泄露.

针对某些数据集中不同属性对类别产生的影响程度有所不同,提出了属性权值的概念,用于改进C4.5决策树算法.通过计算得到不同属性在分类过程中对类别的重要程度,分别赋予不同属性不同的权值,以不同权值计算属性信息增益率,从而找到最佳决策属性;同时将算法运行在HDFS集群,通过Hadoop平台控制多台计算机同时处理待分类数据集,以并行的方式构造决策树.实验结果表明,改进的C4.5算法在处理不同属性对分类结果影响程度不同的数据时比传统C4.5算法具有更高的准确率,并且由于程序并行运行,能够更加高效地处理大型数据,具有很好的可扩展性.

近些年来,我国各地陆续出台了垃圾分类的政策,垃圾分类也已经逐步成为了一种新的生活方式.但是垃圾分类又存在着分类效率低、分类成本高等问题.针对这些问题提出了一种基于卷积神经网络的垃圾分类系统.通过软硬件相结合的方式实现了垃圾投放检测、垃圾种类识别、垃圾精确投放、结果反馈等功能.对于日常生活垃圾的识别率已达91.7％以上,具有识别率高、分类速度快、方便迭代更新、成本低等优点.

针对智能遥控车的蓝牙通信过程中存在的数据劫取、篡改以及通信双方无身份认证等安全问题,利用密码学相关的技术,设计了一个包含身份认证过程、数据加密过程以及数据校验过程的蓝牙通信模块.以Raspberry Pi(树莓派)为主控模块,搭配其他必要硬件模块,开发出了一台智能遥控车.借助蓝牙4.0技术可实现Android手机对小车的控制.并以此为基础,修改了智能车的蓝牙通信模块,实现了通信双方的身份认证以及交互数据的加解密,可基本保证手机与智能车之间的安全通信.

针对普通CAN总线的安全缺陷,通过对CAN总线通信原理与通信数据帧的利用,提出一种车载CAN总线安全机制,使CAN总线具备身份认证与通信加密的能力,从而达到抵抗重放攻击、防止伪造指令注入的效果.利用试验平台对所提出的CAN总线认证与加密机制进行了测评与分析,结果表明,该机制在表现出良好安全性的同时,在128 MHz的总线频率下,安全CAN总线单帧报文的收发可在6 ms之内完成,适应当下智能网联汽车对CAN总线通信速度的要求.

在线题库软件是可根据考试类型自动出题、自动评分的在线考试管理系统.此软件还可以支持教师和学生手动通过即时消息、电子邮件、论坛发送问题和回答问题.在线考试系统的功能利用XML技术、ASP技术和数据访问技术,设计题库、自动生成试卷,并可阅卷.通过一个远程教育在线测试的试题库,实现Web之间的交换数据.

慕课即"大规模开放式网络课程",将课程免费放到网络上,以兴趣导向,凡是想学习的,都可以进来学,不受时空限制.慕课转变教学模式,以个性化学习为导向,引导学生利用微课程找到适合自己的学习方式.由于教材的基本统一性与学生的自身原因,慕课的教学形式在高等教育中比较适合.结合计算机组成原理MOOC教学实践,从教师教学、学生学习、在线考试等方面构建了MOOC有效学习课堂.

This study proposes an iterative closest shape point (ICSP) registration method based on regional shape maps for 3D face recognition. A neutral expression image randomly selected from a face database is considered as the reference face. The point-to-point correspondences between the input face and the reference face are achieved by constructing the points' regional shape maps. The distance between corresponding point pairs is then minimized by iterating through the correspondence findings and coordinate transformations. The vectors composed of the closest shape points obtained in the last iteration are regarded as the feature vectors of the input face. These 3D face feature vectors are finally used for both training and recognition using the Fisherface method. Experiments are conducted using the 3D face database maintained by the Chinese Academy of Science Institute of Automation (CASIA). The results show that the proposed method can effectively improve 3D face recognition performance.

Learning classroom video is used as a teaching resource for students to learn independently in Large scale open online course teaching method. The possibility of learn classroom video used in classroom teaching is trig-gered. Learning classroom video can help students to adapt to the information environment and improve the ability of information thinking in the digital era. Based on the lack of standardization of scientific teaching design and production technology,this paper expounds the design mode of video teaching in learning class,introduces the software technology to make learning classroom video,and the ability to improve the design and production of video.

Different from the application program,the embedded operating system cannot be directly used in different hard?ware platform. Especially for the variety of electronic products using embedded operating system and the quick update,transplan?tation action on the embedded operating system is also very frequent. Embedded operating system needs to accomplish some codes associated with processor by means of C language and assembly language. The main task is to modify this part codes rela?ted to processor hardware platform for transplanting the embedded operating system into its own hardware platform. μC/OS?Ⅱ is a free open source embedded operating system and its code is simple and easy to understand. The analysis of embedded operat?ing system code and transplantation based on the μC/OS?Ⅱfacilitate the understanding of operating system design method and secondary development.

According to the characteristics of the constrained resource in distributed real-time data mining in the Internet of Things (IOT) environment, a distributed data mining method is researched in such environment. Based on the limits of computing ability, storage ability, battery energy resources, network bandwidth, and the Internet single point failure, the distributed network data mining method is researched, and the adaptive technology and peer-to-peer node method are adopted. The DRA-Kmeans algorithm of data mining based on the K-means algorithm is proposed, and the amount of data communication among the sites to reduce the number of iterations and clustering is reduced. Clustering efficiency is improved, and better clustering results and execution efficiency are achieved.

Because of its specificity of embedded electronic products, software related with hardware needs portability before use. How to design and transplant Bootloader which is the first code of device boot program is particularly important. Bootloader of different kernel development platforms need to modify and write. The embedded software classification, bootstrap program classification, function of Bootloader, program design method is introduced, and the working mechanism of U-Boot and the important method is analyzed.

本文探讨了普通地方院校通过综合性设计性实验提高大学生的动手能力，培养应用型人才。把实验学时由原来的8个学时，增加到20个学时。在实验项目的选择上，让学生自己进行设计、装配和调试。难度由原来的单一功能电路模块，改为微小系统，称为综合性和设计性实验。方法上放弃了原来试验箱的做法，从而大大提高了学生动手能力。

Hydrothermal method is used to prepare Zinc Oxide nano material, with the same reaction time using different silicon substrates. The additional crystal structure of ZnO nanorods is studied by using X ray diffraction (XRD), and the morphology and optical properties of ZnO nanorods spectrum is studied by using scanning electron microscopy (SEM) and researching the reflection spectrum. The results show the products are all ZnO nanorod structure having almost the same structure by the different substrates. In terms of the silicon substrate, crystal orientation, type of impurities and doping concentration have no significant effects on the geometry of the ZnO nanorods. This is important for future solar cell fabrication, having no effects to the growth of ZnO nanorods by the substrates.

An important part of construction root file system of embedded Linux system is built based on S3C6410 processor. Parts of the root file system and file types are introduced. A basic root file system is built using the Busybox toolbox. A method of fast transplant of file system is introduced, providing a reference for file system transplant on other processors.

In order to effectively consider the problems of limited resources of equipment node's memory capacity, processing power, and battery power in the environment of data stream, the method of fast and effective extraction mining knowledge is analyzed. DRA-Kmeans clustering algorithm is proposed on the basis of CluStream algorithm, which combines with RA-Cluster algorithm and introduces the adaptive clustering method and improves CluStream algorithm. The clustering accuracy is increased and clustering effective range is optimized in the case of resource constraints.

In order to get more accurate data, it is very important to remove noise in embedded ECG acquisition system based on Bluetooth wireless communication technology. FIR digital signal filtering algorithm based FPGA-FFIR is proposed to complete ECG filtering. Firstly, Simulation is processed in MATLAB to achieve corresponding filtered waveform coefficient. Then comprehensive design and simulation are processed in Quartus-II. Finally the filter is designed to be downloaded to the FPGA chip and debug. When the algorithm is appropriate to change the filtering software programs and operation parameters of the filter, the filter characteristics can be changed. It is particularly effective for the low-frequency pulse interference and random noise.