首页> 外文会议>IEEE International Conference on Computer and Information Technology >Effect of RObust Header Compression (ROHC) and Packet Aggregation on Multi-hop Wireless Mesh Networks
【24h】

Effect of RObust Header Compression (ROHC) and Packet Aggregation on Multi-hop Wireless Mesh Networks

机译:强大的标题压缩(RoHC)和分组聚合对多跳无线网状网络的影响

获取原文

摘要

This paper evaluates the behavioral effect of RObust Header Compression (ROHC) and packet aggregation in multi-hop wireless mesh networks. ROHC itself shows around 20% achieved rate improvement on a line of wireless link even if as the number of mesh router increases, total achieved rate gradually decreases. Packet aggregation cooperating with ROHC provides 4 to 10 times achieved rate improvement and results in maximum 6 times reduction of end-to-end delay. Even if the ROHC and packet aggregation gives such a promising improvement, we need to identify effect of the processing time of the two schemes on overall wireless mesh network behaviors. To do this work, we measure/use ROHC processing time from Pentium 4 and RouterBOARD 230, and apply the results into NS-2 simulations. The simulation results show high performance improvement from ROHC and packet aggregation deteriorates as a general purpose processor in a mesh router takes much time to deal with the two schemes. This result motivates a hardware design be required to speed up the processing units of the two schemes. Consequently, we propose a hardware system model for ROHC and packet aggregation by using SystemC Hardware Description Language (HDL). From the SystemC results, we conclude the proposed built-in processor design for the two schemes keeps performance improvement by enhancing the low speed processing power of general purpose processors.
机译:本文评估了多跳无线网状网络中鲁棒头压缩(RoHC)和分组聚合的行为效果。 RoHC本身显示大约20%的无线链路率改善即使由于网路路由器的数量增加,总计速率逐渐减小。与ROHC协作的数据包聚合提供了4至10次实现的速率改进,并导致最终延迟减少6倍。即使ROHC和分组聚合给出了如此有希望的改进,我们也需要识别两种方案的处理时间对整体无线网状网络行为的影响。为进行这项工作,我们测量/使用来自Pentium 4和RouterBoard 230的RoHC处理时间,并将结果应用于NS-2模拟。仿真结果显示了RoHC的高性能改善,并且数据包聚合随着网路路由器中的通用处理器而劣化,需要花费很多时间来处理这两种方案。该结果激励硬件设计来加速两种方案的处理单元。因此,我们通过使用Systemc硬件描述语言(HDL)提出了一种用于ROHC和分组聚合的硬件系统模型。从SystemC的结果,我们得出结论:通过加强通用处理器的低转速处理能力所提出的内置处理器设计的两个方案保持性能的提高。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号