关键词: 跳频信号, 参数估计, 密度聚类

Abstract:

In low signal-to-noise ratio scenarios, the time-frequency energy of frequency-hopping signals deteriorates and suffers from residual noise interference, leading to incomplete extraction of time-frequency ridges and limited accuracy in temporal parameter estimation. Thus, a time parameter estimation approach for frequency-hopping signals using time-frequency energy distribution clustering is proposed. Firstly, the Euclidean distances of the position indexes of neighboring time-frequency coefficients in the time-frequency diagram on the time axis are calculated, the inflection points are localized by sorting the distance data sequence and calculating the differences, the distance value corresponding to the inflection point is used as the neighborhood radius of the density-based spatial clustering method of applications with noise. Then, by using the energy density difference between frequency-hopping signals and noise, the clustering algorithm is used to identify and extract energy clusters of signals at the same frequency, and the frequency-hopping instants are preliminarily estimated by the boundary points of the clusters. Finally, the density clustering results are corrected by comparing the estimated values of the frequency-hopping period and the distances between the centroids of the neighboring clusters. Experimental results show that the normalized mean square error of the estimated values of hopping time instants is lower than 0.2 at an signal to noise ratio of ?5 dB. This method addresses the challenge of accurately estimating hopping time instant with low signal-to-noise ratios, providing high-precision and robust technical support for communication reconnaissance and radio monitoring in complex electromagnetic countermeasure environments.

Key words: frequency-hopping signals, parameters estimation, density-based clustering