車聯(lián)網(wǎng)中基于分割行為特征的入侵檢測技術(shù)的設(shè)計(jì)與實(shí)現(xiàn)

2萬字
自己原創(chuàng)的畢業(yè)論文，已經(jīng)通過校內(nèi)系統(tǒng)檢測，重復(fù)率低，僅在本站獨(dú)家出售，大家放心下載使用

摘要 隨著車聯(lián)網(wǎng)研究的進(jìn)展和其潛在的應(yīng)用前景，車聯(lián)網(wǎng)的安全問題正逐漸成為人們關(guān)注的焦點(diǎn)。由于車聯(lián)網(wǎng)中節(jié)點(diǎn)高速移動、無線信道質(zhì)量不穩(wěn)定等影響，產(chǎn)生了原本在互聯(lián)網(wǎng)絡(luò)中不存在的安全問題，使得原本有效的安全措施失效。為了保證車聯(lián)網(wǎng)的安全、穩(wěn)定、高效的運(yùn)行，解決各種各樣的車聯(lián)網(wǎng)安全問題，僅依賴一些傳統(tǒng)防御技術(shù)，如防火墻、身份認(rèn)證、數(shù)據(jù)加密等一般被動防御方法已經(jīng)不能做到完全抵御入侵，為此，有必要在車聯(lián)網(wǎng)中引入入侵檢測技術(shù)(IDS)。
本文針對車聯(lián)網(wǎng)的特性，致力于建立適合車聯(lián)網(wǎng)環(huán)境的入侵檢測技術(shù)。首先，介紹了計(jì)算機(jī)網(wǎng)絡(luò)中入侵檢測的相關(guān)概念，發(fā)現(xiàn)了現(xiàn)有入侵檢測技術(shù)存在不足，并通過分析了神經(jīng)網(wǎng)絡(luò)的相關(guān)知識，發(fā)現(xiàn)神經(jīng)網(wǎng)絡(luò)所固有的自適應(yīng)、自學(xué)習(xí)等能力很適合入侵檢測系統(tǒng)的要求，且能夠彌補(bǔ)傳統(tǒng)入侵檢測技術(shù)的不足。然后，還對BP神經(jīng)網(wǎng)絡(luò)以及BP學(xué)習(xí)算法進(jìn)行了研究，發(fā)現(xiàn)BP神經(jīng)網(wǎng)絡(luò)具有的特性很符合車聯(lián)網(wǎng)對于入侵檢測的需求。
由于車聯(lián)網(wǎng)中的數(shù)據(jù)要求高可靠性和及時性，則車聯(lián)網(wǎng)中的入侵檢測需要快速地檢測出入侵行為。為此，將車聯(lián)網(wǎng)中的事件的行為特征分割成幾個特征子集，分別輸入到幾個神經(jīng)網(wǎng)絡(luò)中同時進(jìn)行檢測。在此基礎(chǔ)上，本文在車聯(lián)網(wǎng)的系統(tǒng)模型上設(shè)計(jì)了一種基于分割行為特征的入侵檢測技術(shù)方案。該方案給出了一個基于分割行為特征的入侵檢測系統(tǒng)的總框架，并就框架中各模塊的原理和實(shí)現(xiàn)給予了詳細(xì)的介紹。
最后，利用交通仿真軟件VanetMobiSim搭建道路仿真場景，模擬道路上車輛節(jié)點(diǎn)的運(yùn)動軌跡，并將生成的仿真數(shù)據(jù)作為基準(zhǔn)數(shù)據(jù)集。隨機(jī)抽取數(shù)據(jù)集中的部分?jǐn)?shù)據(jù)作為訓(xùn)練集和測試集,并將數(shù)據(jù)作為入侵檢測系統(tǒng)的輸入，從而檢測出方案的性能。
關(guān)鍵詞：車聯(lián)網(wǎng) 入侵檢測 神經(jīng)網(wǎng)絡(luò) BP學(xué)習(xí)算法

Design and implementation of Intrusion detection technology based on segmentation behavior characteristics in VANET
Abstract With the progress of the VANET research and its potential application prospect, the VANET security problem is becoming the focus of attention. As the VANET nodes in high-speed mobile and wireless channel quality of instability and other affect, resulting in a safety problem originally does not exist in the network, making the original effective security measures fail. In order to ensure safe, stable and efficient operation of vehicular ad-hoc network, to solve all kinds of the vehicular ad-hoc network security issues, only rely on traditional defense technologies such as firewalls, identity authentication, data encryption and other methods have been generally passive defense can not be completely resist invasion, Therefore, it is necessary to introduce intrusion detection technology in the VANET.
In this paper, according to the characteristics of the VANET, is committed to establish suitable for the VANET environment of intrusion detection technology. First of all, Introduces the related concepts about intrusion detection, found the shortcomings of existing intrusion detection technology, and through the analysis of the relevant knowledge of the neural network, found that neural network of adaptive and self-learning ability is very suitable for the requirement of the intrusion detection system, and to be able to make up for the inadequacy of the traditional intrusion detection technology. Then, also on the BP neural network and BP learning algorithm is studied, and found that the characteristics of BP neural network has very accord with VANET demand for intrusion detection.
Due to the data in the VANET requires high reliability and timeliness, the intrusion detection in the VANET need to quickly detect the intrusion behavior. So, the features of events in the VANET divided into several subsets, input into several neural network respectively for testing at the same time. On this basis, the paper design a kind of intrusion detection program based on segmentation behavior characteristics on the system model of the VANET. The program gives an intrusion detection system based on segmentation behavior characteristics of total framework, and on the principle and implementation of each module framework to give a detailed introduction.
Finally, by using traffic simulation software VanetMobiSim build road simulation scenarios, simulated trajectories of vehicles on the road nodes, and the generated simulation data as a baseline data set. Part of the data randomly selected as the training data set and test set, and the data as the input of intrusion detection system, so as to detect the performance of the program.
Key words: VANET Intrusion detection Neural network BP learning algorithm

目 錄
第一章、緒論 1
1.1、課題研究背景及意義 1
1.2、國內(nèi)外的研究現(xiàn)狀及存在問題 2
1.3、本文的主要研究內(nèi)容及組織結(jié)構(gòu) 3
1.3.1 本文的主要研究內(nèi)容 3
1.3.2 本文的組織結(jié)構(gòu) 3
第二章、入侵檢測技術(shù) 5
2.1、入侵檢測技術(shù)概述 5
2.1.1 安全的概念 5
2.1.2、入侵的概念 5
2.1.3、入侵檢測的概念 5
2.2、入侵檢測的發(fā)展歷程 6
2.3、入侵檢測的分類 7
2.3.1 根據(jù)檢測數(shù)據(jù)源分類 7
2.3.2 根據(jù)檢測的分析技術(shù)分類 8
2.4、入侵檢測通用模型 9
2.5、現(xiàn)有入侵檢測技術(shù)的不足 10
2.6、本章小結(jié) 11
第三章、基于..