...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Computer Communications >Cross-layer-based adaptive congestion and contention controls for accessing cloud services in 5G IEEE 802.11 family wireless networks
【24h】

Cross-layer-based adaptive congestion and contention controls for accessing cloud services in 5G IEEE 802.11 family wireless networks

机译：基于跨层的自适应拥塞和竞争控制，用于访问5G IEEE 802.11系列无线网络中的云服务

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The big-data cloud services accessed through 4G LTE-A and 5G WiFi networks need extremely high-speed and highly reliable transmission. However, it exhibits a critical issue that of inconsistently executing congestion control for the global end-to-end connection and contention control for the local wired/wireless links. For instance, in the transport layer (Layer 4 or L4), the reliable TCP adopts ACK-based congestion control to determine the congestion window (denoted by L4_CWND) for the global E2E connection; but, the link layer (Layer 2 or L2) adopts the Truncated Binary Exponential-based (TBE) access control to determine the contention window (denoted by L2_CW) for the local wireless link. Clearly, the congestion and contention control mechanisms for the global end-to-end connection and the local wireless link should cooperate tightly and consistently, but the existing congestion control and contention control are separately operated at different layers. This paper thus proposes a Cross-layer-based Adaptive Congestion Control (namely CACC) for the connection-based transport layer and the link-based media access layer. CACC aims to determine L4_CWND for the end-to-end congestion control, and then sends the cross-layer L4 congestion state to L2 to determine L2_CW_Max and L2_CW consistently. Numerical results demonstrate that CACC outperforms the compared approaches in L2 goodput, L2 collision probability, L2 contention delay, L2 fairness, L4 goodput, L4 fairness, and L4_friendliness. Furthermore, the claims of the determined L4_CWND and L2_CW are supported by mathematical analyses. (C) 2017 Elsevier B.V. All rights reserved.
机译：通过4G LTE-A和5G WiFi网络访问的大数据云服务需要极高的速度和高度可靠的传输。但是，它表现出一个关键问题，即对于全局端到端连接不一致地执行拥塞控制，而对于本地有线/无线链路则执行争用控制。例如，在传输层（第4层或第4层）中，可靠的TCP采用基于ACK的拥塞控制来确定全局E2E连接的拥塞窗口（用L4_CWND表示）。但是，链路层（第2层或第2层）采用基于截断二进制指数（TBE）的访问控制来确定本地无线链路的竞争窗口（用L2_CW表示）。显然，用于全球端到端连接和本地无线链路的拥塞和竞争控制机制应该紧密一致地协作，但是现有的拥塞控制和竞争控制分别在不同的层上进行。因此，本文针对基于连接的传输层和基于链接的媒体访问层提出了一种基于跨层的自适应拥塞控制（CACC）。 CACC的目标是确定用于端到端拥塞控制的L4_CWND，然后将跨层L4拥塞状态发送到L2，以一致地确定L2_CW_Max和L2_CW。数值结果表明，CACC在L2吞吐量，L2碰撞概率，L2竞争延迟，L2公平性，L4吞吐量，L4公平性和L4友好度方面均优于比较方法。此外，所确定的L4_CWND和L2_CW的要求得到数学分析的支持。 （C）2017 Elsevier B.V.保留所有权利。

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