Research on extraction methodbased on adaptive E-pulse polecharacteristic parameters

In order to improve the performance of the traditional E-pulse pole extraction algorithm and provide guidance for rapid decision-making in complex environments, this paper conducts research on the adaptive e-pulse pole extraction algorithm. The time-domain echo data of the standard slender rod is obtained through the FEKO electromagnetic simulation software. The poles are extracted by using the matrix pencil method. The initial E-pulse is constructed according to the principle of E-pulse generation. The decision to optimize pole data is based on whether the normalized energy of the convolution response between the E-pulse and the time-domain echo data exceeds a preset threshold. The gradient descent algorithm is adopted to optimize the extracted poles until the normalized energy meets the preset threshold requirements. By analyzing the time-domain echo data in different excitation directions, it is found that there are differences in the amplitude of the time-domain echo corresponding to different incident angles. The oscillation amplitude of the time-domain echo data gradually decreases to zero over time. The effectiveness of the pole extraction algorithm is proved by analyzing the consistency between the extracted poles at different incident angles and the theoretical poles. By analyzing the difference in the response energy identification number after convolution of E-pulse and echo data constructed by rectangular basis functions and trigonometric basis functions, it is found that the rectangular basis function has a more positive impact on improving the target identification effect than the trigonometric basis function. By analyzing that the energy identification number gradually decreases with the number of iterations, the effectiveness of the gradient descent algorithm is proved.