Wigner—Ville分布是线性调频(LFM)信号参数估计方法中的常用方法,但其缺点是对多分量LFM信号有严重的交义项。文中以稀疏分解方法为基础,利用匹配追踪(MP)算法将微弱多分量LFM信号在过完备原子库上进行分解,由分解得到的原子参数可以估计出各个LFM信号的起始频率和调频斜率,从而实现了微弱多分量LFM信号的参数估计。仿真验证了该方法的有效性。
The Wigner-Ville distribution is a common method for estimating LFM signal parameter. However, the method is not suitable for multicomponent LFM signals due to crossterms in the Wigner-Ville distribution. Therefore, based on the method for the sparse decomposition, weak multicomponent LFM signals are decomposed on the over-complete dictionary of atoms by the matching pursuit (MP) algorithm. The original frequency and LFM rate parameters of LFM signals can be estimated by parameters of MP decomposed atoms. The simulation result demonstrates the effectiveness of the method.