摘要
针对接收到的多个未知任何先验参数的跳频信号,提出一种先分离各个信号再对其分别进行时频分析来估计跳频参数的方法.首先采用特征矩阵联合近似最优化(JADE)算法分离跳频信号,再利用多窗口重叠的平滑伪Wigner Ville分布(SPWVD)来估计出跳频信号的跳周期(hop duration)、定时偏差(time offset)和跳频频率(跳频图案)等参数.通过将多个时频分析的窗口重叠来克服时频分析中交叉项的影响.仿真结果表明,该估计算法显著提高了估计的准确度和可靠性.
This work proposed a method for estimating the parameters of multiple received frequency-hopping (FH) signals without any a priori knowledge. First the unknown signals were separated by joint approximate diagonalisation of eigen-matrices (JADE) algorithm. Then the method used the smoothed pseudo Wigner-Ville distribution (SPWVD) of each separated FH signal to estimate its transmission parameters such as hop duration, time-offset and hopping frequency pattern. By using multiple overlapped SPWVD windows, the effect of cross-term interference in the time-frequency analysis was minimized. Simulation results showed that the new estimation algorithm greatly improved the accuracy and reliability of parameter estimation.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2005年第4期465-470,共6页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金资助项目(60002003).
