首页> 外文期刊>Computer networks >A kernel module for pulse-coupled time synchronization of sensor networks
【24h】

A kernel module for pulse-coupled time synchronization of sensor networks

机译:用于传感器网络脉冲耦合时间同步的内核模块

获取原文
获取原文并翻译 | 示例
           

摘要

The biologically-inspired synchronization paradigm of pulse-coupled oscillators has received increased attention in the communications and sensor network communities as an appealing alternative to traditional packet-based synchronization strategies. Its inherent scalability, simplicity, and decentralized nature make pulse-coupled synchronization an attractive choice for time synchronization in wireless ad-hoc networks. However, in most current implementations, the pulse-coupled synchronization algorithm is coded in the application layer, which makes its performance susceptible to CPU processing load variations. Implementation of the pulse-coupled synchronization strategy in the physical layer is also reported, which however, is subject to difficulties in migration between different platforms. In this paper, we present the design, implementation, and evaluation of the pulse-coupled synchronization strategy as a Linux kernel module. Our goal is to leverage the high portability and prioritized CPU access of kernel modules to 1) reduce the influence of disturbances from application layer programs on the synchronization performance; and 2) simultaneously make the synchronization strategy easily installable in Linux-based sensor networks. Both lab experiments and field tests were conducted to confirm the effectiveness of the results. (C) 2017 Elsevier B.V. All rights reserved.
机译:作为传统的基于分组的同步策略的一种有吸引力的替代方法,脉冲耦合振荡器的生物学启发式同步范例已在通信和传感器网络社区中受到越来越多的关注。它固有的可扩展性,简单性和分散性使脉冲耦合同步成为无线自组织网络中时间同步的诱人选择。但是,在大多数当前实现中,脉冲耦合同步算法在应用程序层中编码,这使其性能容易受到CPU处理负载变化的影响。还报告了物理层中脉冲耦合同步策略的实现,但是,这在不同平台之间迁移时会遇到困难。在本文中,我们介绍了作为Linux内核模块的脉冲耦合同步策略的设计,实现和评估。我们的目标是利用内核模块的高可移植性和优先的CPU访问权限来实现以下目的:1)减少来自应用层程序的干扰对同步性能的影响;和2)同时使同步策略易于安装在基于Linux的传感器网络中。进行了实验室实验和现场测试,以确认结果的有效性。 (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