本文關鍵詞： 轉發(fā)策略 車聯(lián)網 NDN網絡 NS-3　出處：《吉林大學》2017年碩士論文　論文類型：學位論文

【摘要】：汽車在人們的生活中一直扮演著非常重要的角色,并且由于物聯(lián)網的快速發(fā)展和人們對于汽車的需求從單一的駕駛功能向多元化服務的轉變,因此車聯(lián)網的概念被提出,而車聯(lián)網的核心,在于如何更高效率的獲取數(shù)據上。IP網絡由于自身的局限性,可以預見,在將來很難完美地支持車聯(lián)網所提出的各種技術要求。而NDN網絡獨特的路由方式,可以很好的適用于車聯(lián)網通信。NDN網絡只關心獲取的數(shù)據正確與否,而不關注如何獲得數(shù)據和從哪里獲得數(shù)據。而且NDN網絡具有路由緩存功能,數(shù)據可以按照不同的替換策略存儲到路由中,所以數(shù)據可能存儲在多個路由節(jié)點中,大大提高了數(shù)據的獲取效率,并且這種數(shù)據的分布方式也符合車聯(lián)網通信方式。在車聯(lián)網下,車輛在運行過程中,可以直接從鄰居車輛中得到需要的數(shù)據,而不必從數(shù)據的源頭獲得,從而減少了路由轉發(fā)次數(shù),節(jié)省了時間,提高了通信效率。但NDN網絡仍需要解決很多問題。一方面,洪泛廣播策略會導致大量的信息丟失和過高的重傳開銷,這給網絡造成了很大的負擔。另一方面,由于車輛位置的不斷變動,導致拓撲結構不斷的變化,車輛可能會處于斷網的情況,因此降低了重傳效率。本文首先提出了基于距離的轉發(fā)策略,根據節(jié)點到達最終目標地點距離的不同,設置不同的轉發(fā)等待時間。距離越短,轉發(fā)等待時間越短。當某個節(jié)點在等待廣播的過程中,收到了來自其他節(jié)點的相同的興趣包,說明有節(jié)點已經廣播了此興趣包,因此不再重復廣播。這個策略很好地解決了洪泛廣播和興趣包命中率問題,但平均延遲較高。本文又提出了“受歡迎度”這個概念,“受歡迎度”代表了每個節(jié)點所攜帶數(shù)據受歡迎的程度。從車輛節(jié)點所攜帶的數(shù)據這個角度入手,設計了基于“受歡迎度”的轉發(fā)策略,在對興趣包是否重傳的判斷中,“受歡迎度”策略不僅保證了信息傳輸?shù)男?而且成功地降低了平均延遲時間。為了分析本文設計的轉發(fā)策略的性能,利用NS-3仿真平臺下的ndn SIM模擬器,仿真了真實的道路環(huán)境和物理環(huán)境,模擬了車輛的運行軌跡和通信。實驗結果表明,在NDN網絡的V2V通信下,本文所設計的轉發(fā)策略比原有的轉發(fā)策略更好地適用于車輛之間的通信。
[Abstract]:Automobile has always played a very important role in people's life, and because of the rapid development of Internet of things and the change of people's demand for automobile from single driving function to diversified service, the concept of car networking has been put forward. The core of vehicle networking is how to obtain data more efficiently. IP network, due to its own limitations, can foresee that it will be difficult to support all kinds of technical requirements of vehicle networking perfectly in the future. However, NDN network has a unique routing mode. It can be well applied to the vehicle-networking communication. NDN network only cares about whether the obtained data is correct or not, but not about how to get the data and where to get the data. Moreover, the NDN network has routing caching function. Data can be stored in routing according to different replacement policies, so data may be stored in multiple routing nodes, which greatly improves the efficiency of data acquisition. And the distribution of the data is in line with the vehicle network communication mode. Under the vehicle networking, the vehicle can get the needed data directly from the neighbor vehicle during the running process, without having to obtain the data from the source of the data. Thus reducing the number of routing forwarding, saving time and improving communication efficiency. However, NDN network still needs to solve many problems. On the one hand, flood broadcast policy will lead to a large number of information loss and high retransmission overhead. This places a heavy burden on the network. On the other hand, because of the continuous changes in the location of the vehicle, resulting in continuous changes in the topology, the vehicle may be in a situation where the network is cut off. Therefore, the efficiency of retransmission is reduced. Firstly, a distance-based forwarding strategy is proposed in this paper. According to the different distance between the nodes to the final destination, different forwarding waiting times are set. The shorter the distance is, the shorter the forwarding time is. The shorter the forwarding wait time. When a node receives the same packet of interest from other nodes while waiting for broadcast, indicating that some node has already broadcast the packet of interest, This strategy is a good solution to the problem of flooding and interest packet hit ratio. But the average delay is high. This paper also proposes the concept of "popularity", which represents the degree of popularity of the data carried by each node, starting from the point of view of the data carried by the vehicle node. In this paper, we design a forwarding strategy based on "popularity". In judging whether interest packets are retransmitted or not, the "popularity" strategy not only ensures the efficiency of information transmission, but also improves the efficiency of information transmission. In order to analyze the performance of the forwarding strategy designed in this paper, the real road environment and physical environment are simulated by using the ndn SIM simulator based on NS-3 simulation platform. The simulation of vehicle trajectory and communication shows that the proposed forwarding strategy is more suitable for vehicle communication under V2V communication in NDN network than the original forwarding strategy.
【學位授予單位】：吉林大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：U495;TN929.5

【相似文獻】

相關碩士學位論文 前1條

1 李長宇;基于NDN的車聯(lián)網數(shù)據轉發(fā)策略研究[D];吉林大學;2017年

，

本文編號：1552503