首页--工业技术论文--自动化技术、计算机技术论文--计算技术、计算机技术论文--计算机的应用论文--信息处理(信息加工)论文--模式识别与装置论文

Network Data Security for the Detection System in the Internet of Things with Deep Learming Approach

abstract第3页
1 Introduction第6-10页
    1.1 Motivation第7页
    1.2 Problem Statement第7页
    1.3 Purpose, Scope, and Contribution第7-9页
    1.4 Research methodology used第9-10页
2 Background第10-35页
    2.1 General knowledge of ANN第10-25页
        2.1.1 Model for an ANN第13-17页
        2.1.2 Modes of behavior第17-18页
        2.1.3 Classical ML第18-20页
        2.1.4 Structure of ANNs or Multilayered Networks第20-22页
        2.1.5 Formation of a neural network第22-24页
        2.1.6 Evaluation Metrics第24-25页
    2.2 Deep networks neural第25-30页
        2.2.1 Generality of deep neural networks第25-26页
        2.2.2 Feedforward Networks第26-27页
        2.2.3 Feature Introspection第27页
        2.2.4 Random Forest Classifier第27-28页
        2.2.5 Gated Recurrent Unit第28-29页
        2.2.6 Hyper parameters used第29-30页
    2.3 Internet of Things第30-35页
        2.3.1 What is the IoT approaching?第30-31页
        2.3.2 How to secure the Internet of Things第31-32页
        2.3.3 The Importance of Network Security with Io T第32-33页
        2.3.4 Intrusion Detection System第33-35页
3 Related work第35-48页
    3.1 Recurrent Neural Networks第35页
    3.2 Gated Recurrent Unit第35-36页
    3.3 System Architecture第36-37页
    3.4 Dataset第37-38页
    3.5 Introduction第38页
    3.6 Key features for IoT solutions第38-41页
        3.6.1 Light-Weight第39-40页
        3.6.2 Multi-Layered第40页
        3.6.3 Longevity第40-41页
    3.7 Architecture第41-42页
    3.8 IDS - Datasets第42-43页
    3.9 Data Flow第43-44页
    3.10 Experimental Settings第44-48页
        3.10.1 Data Preparation第44-45页
        3.10.2 Feature Engineering第45页
        3.10.3 Hardware and Software used第45-46页
        3.10.4 Algorithm of random classification of the forests第46-48页
4 Implementation and Experimental Results第48-58页
    4.1 Feature Selection第48-51页
    4.2 Evaluation Metrics第51-56页
        4.2.1 Performance results of all classifiers with a hidden layer第52-54页
        4.2.2 Classifier performance results with application layer第54页
        4.2.3 Performance Results of the Transport Layer Classifier第54-55页
        4.2.4 Network layer classifier performance results第55-56页
    4.3 Comparing results produced by classifiers and their layers第56-57页
    4.4 Final comparison of the results第57-58页
5 Conclusion of work第58-59页
6 References第59-62页
ACKNOWLEDGEMENT第62页

论文共62页,点击 下载论文
上一篇:一维光子晶体波导与微腔的控光特性及传感应用研究
下一篇:基于深度学习的工业过程控制系统异常捡测技术的研究