本文關鍵詞： 車輛自組織網(wǎng)絡 交通態(tài)勢感知 廣播方法 分布式傳輸　出處：《電子科技大學》2015年博士論文　論文類型：學位論文

【摘要】：隨著經(jīng)濟的迅速發(fā)展,私家車保有量急劇增長,雖然城市交通路網(wǎng)也得到了極大提升,但汽車保有量遠遠超過了路網(wǎng)的承載能力�，F(xiàn)有城市智能交通系統(tǒng)的局部信息采集和集中處理與分發(fā)機制,已不能適應交通流的大規(guī)模聚集和高突發(fā)性,交通指揮控制嚴重滯后,高峰期仍需大量人工干預,無法從根本上緩解城市交通擁堵問題。為此,急需突破現(xiàn)有智能交通系統(tǒng)框架,建設一種新型智能交通控制系統(tǒng),及時獲得并發(fā)布高實時性交通態(tài)勢給所轄或鄰居區(qū)域,為車輛主動進行擁堵避免和分流提供依據(jù),從而在有效緩解交通擁堵的同時極大地提高道路網(wǎng)絡的運行效率。圍繞這一課題,論文所在項目組提出了一種基于車輛間通信的分布式智能交通系統(tǒng)框架,其中車輛既是交通態(tài)勢的感知節(jié)點、態(tài)勢信息的傳輸(中繼)節(jié)點,又是態(tài)勢信息的接收節(jié)點和最終受益者。其主要關鍵技術包括車輛定位、交通態(tài)勢信息獲取/融合/傳輸、交通流大數(shù)據(jù)分析及其應用(如實時交通地圖更新與發(fā)布等)。本論文先給出分布式智能交通系統(tǒng)架構(gòu),然后針對交通態(tài)勢信息感知與傳輸技術進行研究,具體包括:第一,研究路面上車輛定位、群移動速度獲取和交通狀態(tài)感知方法和技術。首先,針對現(xiàn)有GPS等車輛定位方法不能定位到車輛所在車道、信號易受遮擋等缺陷,提出了一種基于RFID的路面移動車輛主動定位方法(RAVP,RFID based Active Vehicle Positioning),并對RFID標簽的安裝部署方案、定位性能和成本等進行了詳細分析。其次,考慮群移動速度是表征車流的重要參數(shù)之一,給出了一種基于聚簇的群速度感知方法。最后,當獲取了準確的位置信息和群移動速度后,結(jié)合模糊邏輯理論中的模糊映射函數(shù)關系得到交通擁堵等級。提出了一種基于VANETs的交通暢通-擁堵狀態(tài)感知方法(Tra SD-VANET,Traffic Situation Detection method using Vehicular Ad-hoc NETworks)。在車輛密度不同的交通場景中,把Tra SD-VANET與現(xiàn)有方法進行仿真對比,Tra SD-VANET方法不僅可以動態(tài)地感知交通狀態(tài),而且具有較高的準確度,該感知結(jié)果用于交通疏導,可提高車輛的平均行駛速度,減少停等車輛數(shù)目,縮短等待時間,提高了路網(wǎng)效率。第二,節(jié)點非均勻分布和高速移動使VANETs具有拓撲捷變性、網(wǎng)絡節(jié)點突發(fā)性地大規(guī)模聚集性和稀疏節(jié)點密度時的網(wǎng)絡連接通斷性等。針對VANETs的網(wǎng)絡特性帶來的廣播沖突、網(wǎng)絡空洞等問題,分析了在VANETs中傳播信息時的數(shù)據(jù)包成功投遞率和多跳連通性。首先,為分析2-D VANETs中廣播的數(shù)據(jù)包成功接收率,建立了十字路口場景,分析了隱藏終端、并發(fā)節(jié)點、廣播覆蓋范圍、載波偵聽范圍、鄰居的干擾范圍等因素對數(shù)據(jù)包接收的影響,推導了數(shù)據(jù)包的成功接收概率與上述因素間的關系式,仿真驗證了分析結(jié)論。該結(jié)論為后續(xù)多跳廣播的性能評估及應用需求提供了理論依據(jù)。其次,從新的視角分析多跳傳輸?shù)倪B通性,研究了網(wǎng)絡空洞在VANETs中的產(chǎn)生概率和分布率,給出了采用SCF路由算法解決網(wǎng)絡空洞問題時數(shù)據(jù)包成功傳輸?shù)臅r間延遲表達式,并分析了時間延遲在不同車輛環(huán)境下的分布,為所提廣播方法設定攜帶時長提供了參考。第三,為了應對信息傳播過程中面臨的廣播沖突、網(wǎng)絡過載、網(wǎng)絡空洞等問題,提出三種傳播方法。首先,針對網(wǎng)絡節(jié)點的快速移動帶來的鄰居信息快速失效問題,以及高節(jié)點密度環(huán)境下頻繁地廣播信標帶來的廣播沖突和網(wǎng)絡過載問題,提出了一種自適應調(diào)節(jié)信標廣播間隔的方法(SDMB,Self-Decision-Making Beaconing)。在該方法中,節(jié)點根據(jù)自身及鄰居節(jié)點的活躍度,確定信標廣播的時間間隔,以改善網(wǎng)絡性能。結(jié)果表明,與IEEE802.11p的WAVE模型中采用的固定速率廣播beacon的方法相比,SDMB方法有較高的信標廣播成功率和較低的網(wǎng)絡負載率。其次,針對由車輛節(jié)點的高速移動和非均勻分布導致的網(wǎng)絡空洞問題,提出了一種先應式網(wǎng)絡空洞發(fā)現(xiàn)策略,在此基礎上提出時間受限的攜帶貪婪前傳方法(CGFB,Carry Greedy Forward Broadcast)。該方法中的空洞發(fā)現(xiàn)策略是依據(jù)精確的位置信息建立鄰居排序表,通過查詢表格中的鄰居存無和節(jié)點間的幾何關系發(fā)現(xiàn)網(wǎng)絡空洞,再通過攜帶貪婪前傳,有效解決由網(wǎng)絡空洞導致的鏈路中斷問題,提高了成功傳輸數(shù)據(jù)包的概率。同時,根據(jù)網(wǎng)絡空洞的分布概率設定攜帶時長,以降低對緩存資源的消耗。結(jié)果表明,CGFB算法在建立連通路徑的成功率和節(jié)省存儲資源方面優(yōu)于GF和SCF算法。最后,針對事故信息的影響特點以及特定目標區(qū)域?qū)π畔V播的需求,提出了一種空時受限的交通態(tài)勢信息傳播方法(STLIB,Space-Time Limited Infection Broadcast)。該方法在限定的時間和目標區(qū)域內(nèi)及時廣播交通信息,告知后續(xù)車輛及時分流繞避交通事件/事故發(fā)生路段。結(jié)果表明,使用STLIB方法后,車輛排隊長度顯著縮短,有效緩解了交通擁堵。
[Abstract]:With the rapid development of economy, the rapid growth of private cars, although the city traffic has been greatly improved, but the cars far exceeded the carrying capacity of the road network. The existing city intelligent transportation system local information collection and centralized processing and distribution mechanism, already can not adapt to the traffic flow in large-scale aggregation and bursty. Traffic control is seriously lagging behind, the peak still needs a lot of manual intervention, can not alleviate the city traffic congestion problem fundamentally. Therefore, the urgent need to break the existing framework of intelligent transportation systems, the construction of a new intelligent traffic control system, timely access to and release the high real-time traffic situation to the jurisdiction or neighborhood, for vehicle active congestion to avoid the shunt and provide the basis, thus greatly improve the efficiency of the road network at the same time effectively alleviate traffic congestion. Around this topic, the The project team proposed a framework of distributed intelligent transportation system based on inter vehicle communication, the vehicle is aware node traffic situation, transmission state information (relay) node, and is state information receiving node and the ultimate beneficiaries. Its key techniques include vehicle location, traffic situation information acquisition / fusion / transmission traffic flow, big data analysis and Its Applications (such as real-time traffic map updating and Publishing). This paper first presents a distributed intelligent transportation system architecture, and then according to the traffic situation information sensing and transmission technology research, including: first, vehicle positioning on the road, moving speed and traffic state group gets perception method and technology. First, according to the existing GPS vehicle positioning method cannot navigate to where the vehicle lane, the signal is susceptible to occlusion and other defects, proposed a road mobile vehicle based on RFID Active positioning method (RAVP, RFID based Active Vehicle Positioning), and install the deployment scheme of the RFID tag, the positioning performance and cost are analyzed in detail. Secondly, considering the group moving speed is one of the important parameters of traffic characterization, presents a group velocity sensing method based on clustering. Finally, when the the position information and the moving speed of accurate group, combined with the fuzzy mapping function relationship logic theory in the traffic congestion level. Proposed a method for congestion aware VANETs traffic - based (Tra SD-VANET, Traffic Situation Detection method using Vehicular Ad-hoc NETworks). The density of vehicles in different traffic scenarios, the Tra SD-VANET simulation comparison with the existing methods, Tra SD-VANET method can not only dynamically perceive the traffic state, but also has high accuracy, the sensing results For traffic, the average speed of vehicles can be enhanced, and reduce the number of vehicles, shorten the waiting time, improve the efficiency of road network. Second, non uniform node distribution and high speed mobile VANETs network node topology with Czech degeneration, sudden large-scale aggregation and sparse node density when the network connection continuity. According to the characteristics of VANETs radio network conflict caused by network hole. Analysis of the dissemination of information in VANETs data packet delivery success rate and multi hop connectivity. First of all, for the analysis of 2-D broadcasting in VANETs packet reception success rate, a crossroads scene, analyzes the hidden terminal, concurrent node broadcast the coverage, the carrier sensing range, the range of interference factors of neighbor effect on data packet reception, successful packet reception probability and the above factors were deduced, simulation The conclusions of the analysis. The conclusion provides a theoretical basis for the performance evaluation and application of the needs of the subsequent multi hop broadcast. Secondly, connectivity analysis of multi hop transmission from a new perspective, studies the network empty in VANETs probability and distribution rate, given the packet transmission time delay network problem the expression of SCF routing algorithm, and analyzes the time delay distribution in different vehicle environment, for the broadcast setting method when carrying long reference. Third, in order to cope with the broadcast conflict in the process of information transmission, network overload, network problems such as empty, put forward three kinds of communication methods. Firstly, the problem of fast according to the rapid failure of the mobile network node from the neighbor information, and high node density environment frequently brings conflict and radio broadcasts a beacon network overload problem, we propose an adaptive Method of regulating beacon broadcast interval (SDMB, Self-Decision-Making, Beaconing). In this method, node and neighbor nodes according to their activity, determine the beacon broadcast time interval, in order to improve the network performance. The results show that compared with the fixed rate of broadcast beacon using the WAVE model in IEEE802.11p, the SDMB method has the load rate radio beacon high success rate and low network. Secondly, according to the vehicle by the high-speed mobile nodes and non uniform distribution caused by network hole problem, proposed a proactive network hole discovery strategy based on time constrained greedy methods carry prequel (CGFB, Carry Greedy Forward Broadcast). In the method of empty found method is on the basis of accurate location information of neighbor table, through the query form the neighbor nodes and no deposit geometry relationship found in the network Empty, then carry the greedy prequel, effectively solve the problem of interrupt by the link network empty result, improve the probability of successful packet transmission. At the same time, according to the probability distribution network empty set carry long, in order to reduce the cache resource consumption. The results show that the CGFB algorithm in the establishment of the communication path and the success rate of saving the storage resource is better than GF and SCF algorithm. Finally, the influence of characteristics of accident information and target specific areas of demand for information broadcasting, the traffic situation information dissemination method of space-time constrained (STLIB, Space-Time Limited Infection Broadcast). This method in time and the target area in the limited time radio traffic information. Inform the following vehicle timely triage around to avoid the traffic incident / accident road. The results show that using the STLIB method, the vehicle queue length was significantly shortened, effectively alleviate traffic Traffic jam.

【學位授予單位】：電子科技大學
【學位級別】：博士
【學位授予年份】：2015
【分類號】：U491
，

本文編號：1553904