首页> 外文期刊>Aerospace and Electronic Systems, IEEE Transactions on >Performance improvement in DTNs by packet size optimization
【24h】

Performance improvement in DTNs by packet size optimization

机译:通过数据包大小优化提高DTN的性能

获取原文
获取原文并翻译 | 示例
           

摘要

In space communication environments, most studies in delay tolerant networks (DTNs) adopt experimental approaches and they have shown that the utilization efficiency of link bandwidth can be affected by changing packet sizes at the bundle layer and the convergence layer. However, the lack of theoretical work makes it difficult to find the optimal packet sizes in general scenarios. In this paper, we focus on this important issue by performing a theoretical analysis to solve the packet size optimization problems in both one-hop and multihop DTNs. As our analysis is based on the widely used DTN architecture with Bundle Protocol (BP) and Licklider Transmission Protocol (LTP), the theoretical results can be applied to any DTN scenario using this architecture. We formulate packet size optimization problems in DTNs as nonconvex optimization problems. Furthermore, a goodput enhancement algorithm (GEA) is also proposed to find the optimal packet sizes in order to maximize goodput in DTNs. We conduct numerical analysis as well as simulations to validate our theoretical analysis and study the factors that determine the optimal packet sizes for BP and LTP. Numerical results show that GEA achieves significant performance improvement compared with schemes where packet size optimization on BP and LTP is not jointly considered.
机译:在空间通信环境中,大多数对时延容忍网络(DTN)的研究都采用实验方法,它们表明,链路带宽的利用效率会受到捆绑层和会聚层上分组大小的改变的影响。但是,缺乏理论上的工作使得在一般情况下很难找到最佳的数据包大小。在本文中,我们通过进行理论分析以解决一跳和多跳DTN中的数据包大小优化问题,将重点放在这一重要问题上。由于我们的分析是基于广泛使用的具有捆绑协议(BP)和Licklider传输协议(LTP)的DTN体系结构,因此理论结果可以应用于使用此体系结构的任何DTN方案。我们将DTN中的数据包大小优化问题表述为非凸优化问题。此外,还提出了一种吞吐量提高算法(GEA),以找到最佳的数据包大小,以最大化DTN中的吞吐量。我们进行数值分析和模拟,以验证我们的理论分析并研究确定BP和LTP最佳数据包大小的因素。数值结果表明,与不结合考虑基于BP和LTP的数据包大小优化的方案相比,GEA可以显着提高性能。

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