Understanding congestion in IEEE 802.11b wireless networks

Understanding congestion in IEEE 802.11b wireless networks
复制标题

DOI:
10.1145/1330107.1330140
复制
发表时间:
2005-10
期刊:
--
影响因子:
--
通讯作者:
Amit P. Jardosh;K. Ramachandran;K. Almeroth;E. Belding-Royer
Amit P. Jardosh;K. Ramachandran;K. Almeroth;E. Belding-Royer
中科院分区:
其他
文献类型:
--
作者:
Amit P. Jardosh;K. Ramachandran;K. Almeroth;E. Belding-Royer

文献摘要

被引文献

相似文献

无线网络的日益普及导致了大量使用和拥塞的情况。在使用率较高的无线网络中,网络的无线部分是一个主要的性能瓶颈。了解此类网络的无线部分的行为对于确保其稳健运行至关重要。了解这一点还有助于优化网络性能。在本文中,我们使用从第62届互联网工程任务组(IETF)会议上部署的运营、大规模和高使用率的IEEE 802.11b无线网络收集的链路层信息来研究无线网络中的拥塞。我们鼓励使用信道繁忙时间作为信道利用率的直接衡量标准,并展示了如何使用信道利用率以及网络吞吐量和吞吐量来定义高度拥塞、中度拥塞和非拥塞网络状态。我们的研究将网络拥塞及其对链路层性能的影响联系起来。基于这些相关性,我们发现(1)当前的速率自适应实现很少使用2 Mbps和5.5 Mbps的数据速率,(2)使用请求发送/允许发送(RTS-CTS)阻止节点获得对严重拥塞的信道的公平访问,以及(3)作为对拥塞的响应,使用速率自适应对网络性能是有害的。
The growing popularity of wireless networks has led to cases of heavy utilization and congestion. In heavily utilized wireless networks, the wireless portion of the network is a major performance bottleneck. Understanding the behavior of the wireless portion of such networks is critical to ensure their robust operation. This understanding can also help optimize network performance. In this paper, we use link layer information collected from an operational, large-scale, and heavily utilized IEEE 802.11b wireless network deployed at the 62nd Internet Engineering Task Force (IETF) meeting to study congestion in wireless networks. We motivate the use of channel busy-time as a direct measure of channel utilization and show how channel utilization along with network throughput and goodput can be used to define highly congested, moderately congested, and uncongested network states. Our study correlates network congestion and its effect on link-layer performance. Based on these correlations we find that (1) current rate adaptation implementations make scarce use of the 2 Mbps and 5.5 Mbps data rates, (2) the use of Request-to-Send/Clear-to-Send (RTS-CTS) prevents nodes from gaining fair access to a heavily congested channel, and (3) the use of rate adaptation, as a response to congestion, is detrimental to network performance.