Time-Difference-of-Arrival Estimation Algorithms by Employing Cyclostationary Property of LFM Signals

In this letter, two time-difference-of-arrival (TDOA) estimation algorithms are studied by employing the cyclostationary property of LFM signals. First, the multicycle frequency phase TDOA (MCFP-TDOA) estimation algorithm is developed by using the phase difference of the cyclic auto-correlation function of each cycle frequency, followed by an analysis on the phase ambiguity problem. By taking advantage of multiple cycle frequencies and the linear relationship between each cycle frequency, it was found to have smaller estimation error compared with the traditional cyclic phase difference (CPD) method. Second, based on the characteristics of LFM signals in the cyclic frequency domain, the spectral correlation TDOA (SC-TDOA) estimation algorithm is devised, which has good antinoise ability. The improved performances of the proposed algorithms are demonstrated through detailed theoretical analysis and simulations.