跳频信号的相参与非相参积累时频差估计方法

Coherent and non-coherent integration TDOA/FDOA estimation method of frequency-hopping signals

跳频信号载频随时间变化发生跳变,因此跳频信号具有丰富的频域信息,而且其在时域上为连续信号,相比单跳信号也具有更丰富的时域信息。时频差估计精度与信号在时频域上的分布情况以及信号能量和噪声有关。时差估计主要与信号频域分布有关,而频差估计主要与信号时域分布有关。跳频信号时频域信息丰富,多跳相参积累后,时频差参数估计能够充分利用信号的时频域信息,克服单跳信号时频差估计精度不高的问题。针对跳频信号的时频差估计问题,首先分析了单跳基带信号的互模糊函数,再从时差与频差的维度推导了多跳基带信号互模糊函数的相位差异,最后通过频差归一化与相位对齐补偿提出了多跳信号互模糊函数相参积累的时频差估计算法。同时,在不满足相参积累的条件下,分析了跳频信号的非相参时频差估计方法,并理论分析了两种跳频信号时频差参数估计方法性能与信号各参数的关系。仿真结果表明跳频信号相参时频差估计算法性能最优,非相参方法的估计性能其次,单跳信号的估计性能最差,验证了理论性能分析结果的正确性。

The carrier frequency of frequency-hopping(FH)signals varies with time,therefore,the FH signals have broad distribution in frequency.And FH signals are continuous in time domain,which have broader distribution in time domain compared to single hop signal.The estimation accuracy of time difference of arrival(TDOA)and frequency difference of arrival(FDOA)is determined by the frequency distribution or time distribution and energy of signal and noise.The TDOA estimation is mainly depended by the frequency distribution,while the FDOA estimation is mainly depended by the time distribution.FH signals have broad distribution in frequency and time domain,the problem of low estimation accuracy for TDOA and FDOA of single hop signal can be overcome through multi-hop coherent integration to improve the use of information in time and frequency domain.Focus on the TDOA/FDOA estimation problem of FH signals,firstly,the cross ambiguity function(CAF)of single hop baseband signal is analyzed,then,the phase difference of each CAF for the multi hop baseband signal is analyzed through TDOA and FDOA dimensions.Coherent integration TDOA/FDOA estimation method of CAF for multi-FH signals is realized by FDOA normalized compensation and phase compensation.When the coherent integration requirements are not satisfied,the non-coherent TDOA/FDOA estimation method of FH signals is proposed.The relationship between performance of two TDOA/FDOA estimation method for FH signals and each parameter for signal are analyzed theoretically.Simulation results show that the performance of proposed coherent integration TDOA/FDOA estimation method for FH signals is best,and the non-coherent integration method is suboptimal,while that of single hop signal is poorest,which proves the correctness of the theory analysis.