航空学报 > 2014, Vol. 35 Issue (7): 1966-1976   doi: 10.7527/S1000-6893.2013.0526

X射线脉冲星信号时延的实时估计方法

李鹏飞, 徐国栋, 董立珉, 候天蕊   

  1. 哈尔滨工业大学 卫星技术研究所, 黑龙江 哈尔滨 150001
  • 收稿日期:2013-09-16 修回日期:2014-01-20 出版日期:2014-07-25 发布日期:2014-02-13
  • 通讯作者: 徐国栋,Tel.:0451-86414117E-mail:xgdong_61@163.com E-mail:xgdong_61@163.com
  • 作者简介:徐国栋男,博士,研究员,博士生导师。主要研究方向:卫星综合电子系统,通信及导航技术。Tel:0451-86414117E-mail:xgdong_61@163.com;李鹏飞男,博士研究生。主要研究方向:航天器自主导航技术。Tel:0451-86414117E-mail:lpf_365@163.com
  • 基金资助:

    国家“863”计划(2008AA8051602)

A Real Time Estimation Method of Time-delay for X-ray Pulsar Signal

LI Pengfei, XU Guodong, DONG Limin, HOU Tianrui   

  1. Research Center of Satellite Technology, Harbin Institute of Technology, Harbin 150001, China
  • Received:2013-09-16 Revised:2014-01-20 Online:2014-07-25 Published:2014-02-13
  • Supported by:

    National High-tech Research and Development Program of China (2008AA8051602)

摘要:

针对在频域内计算X射线脉冲星信号延时量存在滞后性,进而难以为航天器自主导航提供实时信息的问题,提出将脉冲星信号延时估计转化为时域内标量估计的方法。首先,通过人工神经智能网络获得脉冲星信号的标准轮廓函数作为状态方程;应用粒子滤波算法对脉冲星信号延时量进行实时估计;其次,为了避免标准粒子滤波器中的粒子退化现象,推导并证明了一种新型粒子滤波算法;最后,推导出粒子滤波算法的精度函数,为航天器的导航策略提供参考。以航天器在轨运行中可能遇到的3种情况为背景,验证了所提粒子滤波算法的正确性与有效性。

关键词: 脉冲星, 信号轮廓模型, 人工神经网络, 粒子滤波, 粒子退化

Abstract:

In order to make up for the defect of traditional treatment in the frequency domain, that can not provide enough new information to the autonomous spacecraft navigation system because of long time accumulation, a new kind of particle filter based on the normal particle filter is proposed for estimating the phase of a pulsar signal. Firstly, a mathematical model of the pulsar signal is obtained by artificial neural network, and the model is taken as the system state equation. Secondly, the new particle filter which is proposed in order to solve the sample impoverishment phenomenon is strictly proven. Finally, because of the special features of the system, the function which indicates the output value precision is derived, and it can provide a reference for spacecraft navigation strategy. Several simulation results all show that the filter can guarantee stability and high precision under the conditions which are set for the three cases that the spacecraft may encounter in real situations.

Key words: pulsar, signal profile model, artificial neural network, particle filter, sample impoverishment

中图分类号: