欢迎光临,现在是     设为首页   加入收藏   联系我们
文章摘要
朱振海,沈重.一种基于人工智能-深度学习的无线时钟同步算法[J].海南大学学报编辑部:自然科学版,2018,36(4):.
一种基于人工智能-深度学习的无线时钟同步算法
A Wireless Clock Synchronization Algorithm Based on Artificial Intelligence-Deep Learning
投稿时间:2018-10-13  修订日期:2018-10-31
DOI:
中文关键词: 超宽带;时钟同步算法;人工智能;神经网络;深度学习;TPSN协议
英文关键词: UWB; clock synchronization algorithm; artificial intelligence; neural network;deep learning;TPSN protocol
基金项目:国家自然科学(No.61461017);海南省自然科学基金创新团队项目(No.2017CXTD0004);军委装备发展部预先研究项目(No.6141B010109).
作者单位E-mail
朱振海 海南大学 信息科学技术学院 9580011@qq.com 
沈重 海南大学 信息科学技术学院 592835697@qq.com 
摘要点击次数: 82
全文下载次数: 70
中文摘要:
      UWB TDOA算法定位的准确性很大程度上取决于时钟同步性能,传统TPSN协议没有充分考虑时钟漂移和各种随机时延对时钟同步精度的影响,严重限制时钟同步性能.基于传统TPSN协议基础上,提出一种结合人工智能-深度学习的无线时钟同步算法,利用深度学习不断重复学习的特性,减少误差,抵消时钟漂移的影响和抑制各种随机时延的不确定性.并通过在Hainan EVK RTLS 3.0硬件平台上测试,可知所提算法比传统TPSN协议获得了更高的时钟同步精度,时钟同步性能得到大幅提升.
英文摘要:
      The accuracy of UWB TDOA algorithm positioning largely depends on the clock synchronization performance. However, the traditional TPSN protocol does not fully consider the influence of clock drift and various random delays on the accuracy of clock synchronization, and severely limits the clock synchronization performance. Based on the traditional TPSN protocol, a wireless clock synchronization algorithm based on artificial intelligence and deep learning is proposed. It uses the characteristics of continuous learning of repeated learning in deep learning to reduce errors, counteract the effects of clock drift, and suppress the uncertainty of various random delays. By testing on the Hainan EVK RTLS 3.0 hardware platform, it can be seen that the proposed algorithm obtains higher clock synchronization accuracy than the traditional TPSN protocol, and the clock synchronization performance is greatly improved.
关闭