Internet is playing a more and more important role in shaping our lives, which raises the demand for being connected anytime and anywhere, even while on driving cars. Compared with 3G or 4G networks, roadside APs have the advantages of low cost and high bandwidth. Making use of passing vehicles to carry the requested data for the client can reduce the influence of intermittent connectivity due to APs' sparse distribution. High node mobility, fast topology changes and collision domain overlay those cause the data carried by cooperative vehicles cannot be received totally. Based on previous work, this paper proposes a compensation model of cooper-ative downloading improving system throughput. In highway scenarios, according to the informa-tion of speed and position, a cooperative vehicle estimates the probability of other vehicles en-countering users while meeting, and chooses adapted vehicles to backup partial data aiming at compensation for possible collisions. Simulation results indicate the benefits of the proposed scheme in terms of increasing throughput and reducing delay.%随着人们对互联网依赖性的日益提高,人们希望随时随地能够接入Internet,即使在行驶中的汽车里.相比于3G或4G网络,通过路边接入点(AP)接入互联网具有高带宽低延迟的特点.利用经过的车辆携带用户所需要的数据可以有效地减少由于AP分布稀疏所引起的间歇性连接的影响.然而高速行驶的节点,快速的拓扑变化以及传输碰撞域的重叠将导致协助车所携带的数据不能完全地传递给用户,从而降低了系统的吞吐量.文中在先前工作的基础上提出了一种提高系统吞吐量的协助下载补偿模型.在高速公路场景下,协助车在与同向和对向的其它车辆相遇时根据速度和位置信息预测其与目标车相遇的概率,并选取满足条件的车辆备份携带的数据以弥补因传输碰撞域重叠等因素所引起的部分数据包传送失败的损失,从而达到有效地利用盲区(Dark Area)延伸用户下载区域的目的.实验结果表明该方法显著地提高了下载的吞吐量,减少间歇性连接带来的影响.
展开▼
