2015年VANET网络连接性能深度分析

PDF文件

下载需积分: 5 | 808KB | 更新于2024-08-13 | 107 浏览量 | 举报收藏

立即下载

本文旨在深入探讨车辆自组网络(Vehicular Ad Hoc Networks, VANETs)的网络连接性能，发表在《移动和无线通信》(International Journal of Ad Hoc and Ubiquitous Computing) 2015年第2期。作者Ruifeng Chen、Zhangdui Zhong来自北京交通大学的铁路交通控制与安全国家重点实验室，Chih-Yung Chang来自台湾 Tamkang University 的计算机科学与工程系，以及Bo Ai和Ruisi He也来自北京交通大学的同一实验室。研究的核心内容是建立一个简化但合理的高速公路模型，从个体车辆和全局网络的角度来评估VANETs的网络连通性。论文主要关注三个关键参数： 1. **个体车辆间通信对的条件连通性**：这是衡量一辆车能否与周围其他车辆成功建立通信链接的基础指标。它考虑了距离、速度、干扰等因素对通信质量的影响，是评估网络局部连通性的关键。 2. **全局网络连通性概率**：该参数关注整个VANET系统中车辆之间的连通性情况，它涉及多个车辆之间的交互作用，反映了网络的整体稳定性及信息传播能力。 3. **车辆隔离概率**：这是一个衡量单个车辆与其他车辆断开连接的可能性，这可能由于信号阻挡、频道竞争或其他技术限制导致。降低车辆孤立的概率有助于提高整体网络效率。论文通过计算车辆间的间隔分布函数，结合车辆运动模型，对这些参数进行了深入分析。研究方法可能包括理论推导、仿真建模以及数据驱动的性能评估。通过这些分析，研究者能够提出优化策略，如路由算法或频谱管理措施，以提升VANET的网络连通性和可靠性。这篇论文为理解和改进VANET的网络设计提供了重要的理论依据，对于保障智能交通系统的实时信息共享、车辆安全和道路协同至关重要。随着自动驾驶和物联网技术的发展，车辆自组网络的连通性研究具有长远的实际应用价值。

Performance analysis on network connectivity for vehicular ad hoc networks 69

percolation theory. Some investments focused on the k-

connectivity (Wan et al., 2010), partial connectivity (Haiyan

et al., 2010) or asymptotic connectivity (Honghai et al.,

2005). The measurements of the connectivity performance

includes the critical communication range, the distribution of

the connectivity distance and node population, as well as the

connectivity probability.

Fundamental characteristics of the connectivity in

VANETs were studied in Shioda et al. (2008) with extensive

simulations, and the authors addressed how the road and

vehicle positions would affect the connectivity characteristic.

A new mobility model was presented in Hafeez et al.

(2010) to derive the number of vehicles on the road

and connectivity probability. Two probabilities, the access

probability that a vehicle can access the infrastructure and the

connectivity probability that all the vehicles can connect to

the infrastructure, were derived for vehicular relay networks

in Seh et al. (2011). Reference Youseﬁ et al. (2008) studied

the statistical properties of connectivity in VANETs with the

queuing theory.

The authors in Cheng et al. (2012) discussed the impacts

of the intervehicle spacing distribution on the connectivity of a

vehicular ad hoc network in a highway trafﬁc scenario based on

a new empirical analysis. Reference Zhongjiang et al. (2012)

studied the connectivity in VANETs based on the uniform

stationary distribution without considering the mobility. A

new mobility model was proposed in Hafeez et al. (2013) and

the authors considered how the vehicles speeds would affect

the connectivity and the packet reception rates. In Pitsiladis

et al. (2012), spanning tree theory was adopted to present the

connectivity calculation in ﬁnite wireless multi-hop networks.

A general close form expression for the probability of the

existence of at least one connected spanning tree inside the

network’s graph was derived. Literature (Ren et al., 2011)

evaluated the percolation-based connectivity performance for

the secondary network in a large-scale ad hoc heterogeneous

network, based on the theory of continuum percolation.

Through analysing the existing research results, it is clear

that most of the previous literatures studied connectivity in

VANETs mainly from the macroscopic view, such as node

population and cluster size in the road. The effect of individual

vehicle’s movement on connectivity, which can be regarded

as the microscopic view, is still not clear. Macroscopic view

can be considered as the viewpoint of the entire network and

it mainly focuses on the global properties of the network,

while microscopic view can be regarded as the viewpoint

of a single vehicle or a communication pair of individual

vehicles. In the viewpoint of the macroscopic, connectivity

means the whole network is fully connected. In the viewpoint

of microscopic, connectivity means that there exist some

available paths between the communication pair of vehicles.

Since a great many safety-related services could be

broadcasted only within a small-range zone rather than the

whole network (for instance, the safety messages containing

the lane-changing information or collision avoidance warning

could be delivered to the surrounding vehicles only), the

microscopic properties are of great signiﬁcance to ensure

the reliable safety-related communication in these cases.

Moreover, the inter-vehicle distance is one of the key factors

that affect the microscopic connectivity properties, which is

impacted by the relative movements of the individual vehicles.

Nonetheless, the distributions of the relative movement and

inter-vehicle distance have been paid less attention in most

of the previous works. The previous works analyse the

connectivity properties considering either the vehicle mobility

or the spatial distribution, while in this paper we take both the

relative speed and inter-vehicle distance into account.

1.2 Contribution

The objective of this paper is to study the network connectivity

in a highway mobility model from both the microscopic and

macroscopic view. We focus on the distributions of the relative

speed and inter-vehicle distance, and study how the individual

vehicle’s relative movements affect the microscopic and

the macroscopic connectivity performance. We then employ

three parameters: conditional connectivity probability, global

network connectivity probability, and the vehicle isolation

probability, to investigate the impact of the different system

parameters on the network connectivity.

The conditional connectivity is the probability that a

communication pair can stay connected during a whole data

transmission conditioned on the connection is successfully

built up when the transmission is initialised. In the VANETs

scenario, various types of data are transmitted in the network.

Some small sized data can be transmitted by packaging

into a few or even one packet (for example, the emergency

broadcast), while some other data may need much longer

transmission duration (for example, the video-on-demand

data stream). It is clear that the small sized data has better

adaptivity to the not guaranteed network connectivity, but

the transmission of data for entertainment services may face

frequent link interruptions. Therefore, we will focus on the

effect of conditional connectivity on the data transmission

service like entertainment or other large-data-based ones. In

particular, we will study how the conditional connectivity

would be inﬂuenced by vehicles’ relative mobility and inter-

vehicle initial distance.

Moreover, different from the previous work (Chen

et al., 2012) which only focuses on a communication pair

of individual vehicles, this paper also studies the global

properties of the whole network. Global network connectivity

is the probability that the whole network is fully connected

and an arbitrary vehicle can directly communicate with its

neighbourhood; while vehicle isolation probability means

the probability that a randomly chosen vehicle has no

neighbour and cannot directly communicate with others.

Considering a highly dynamic network topology and the

stringent delay requirements for some of the VANETs

safety-related applications, the delay-critical messages like

collision avoidance warning are supposed to be disseminated

immediately through direct one-hop communication. There

should exist the fully connected paths for the safety-

related message dissemination through the entire network to

guarantee the successful reception of the whole network. In

order to ensure the reliable broadcast for the one-hop safety-

critical services, it is essential that there exists no isolated

vehicle in the network and any arbitrary vehicle can establish

剩余10页未读，继续阅读

weixin_38501610

粉丝: 4

2015年VANET网络连接性能深度分析

无线ad hoc网络多性能指标基本性能边界

基于排的车辆自组织网络的连接概率分析

受用户行为影响的车载自组织网络的连接性能分析

车辆自组织网络仿真研究：性能分析与挑战

检测车辆自组织网络的热路移动性

城市车辆自组织网络中数据传递交叉口的决策

MATLAB源码集锦-车载自组织网络中路边性能及防碰撞算法研究

VANET:在车辆自组织网络 (VANET) 上实施高效的洪泛算法和多点中继 (MPR) 算法，用于半卡车协作自适应巡航控制

VanetMobiSim车辆自组织网络模拟器使用指南

车载自组织网络路由协议深度分析与前景展望

最新资源