中圖分類號： TN914.41 文獻標識碼： A DOI：10.16157/j.issn.0258-7998.211459 中文引用格式： 劉佳敏，趙知勁，葉學義，等. 基于形態(tài)學濾波和時頻譜圖對消的多跳頻信號參數(shù)估計[J].電子技術應用，2021，47(12)：83-88. 英文引用格式： Liu Jiamin，Zhao Zhijin，Ye Xueyi，et al. Morphological filtering and time-spectrogram cancellation based parameters estimation algorithm of multi FH signals[J]. Application of Electronic Technique，2021，47(12)：83-88.
Morphological filtering and time-spectrogram cancellation based parameters estimation algorithm of multi FH signals
Liu Jiamin1，Zhao Zhijin1，2，Ye Xueyi1，Wang Lijun2
1.School of Communication Engineering，Hangzhou Dianzi University，Hangzhou 310018，China； 2.State Key Lab of Information Control Technology in Communication System， The 36th Research Institute of China Electronics Technology Group Corporation，Jiaxing 314001，China
Abstract： Aiming at estimating the parameters of multi-frequency hopping signals in complex electromagnetic environment, a blind estimation algorithm of signals parameters based on multi-scale morphological filtering and time-spectrogram cancellation is proposed. Firstly, considering the characteristic differences of frequency hopping signals, interference signals and noise, multi-scale morphological filtering is used to eliminate the noise, frequency sweep signals and burst signals, the time-spectrogram cancellation is used to remove the fixed-frequency signals. Then the position information of the frequency hopping signals is obtained through the eight-connected domain mark, and then the all-speed frequency hopping signals are separated by the improved K-means clustering algorithm. Finally, the period, hopping time and frequency of multiple frequency hopping signals are estimated according to the parameters of each class cluster. The simulation results show that compared with the estimation algorithm that uses morphological filtering and extracts the time-frequency ridge, the proposed algorithm has higher estimation accuracy under low signal-to-noise ratio, and even though the frequency collision between frequency hopping signals and the fixed frequency signals occurs, the frequency hopping parameters can still be estimated accurately.
Key words : frequency hopping signal；parameter estimation；morphological filtering；time-spectrogram cancellation；clustering
