| Frequency hopping technology is one of the most effective technology in communication anti-jamming technology,combined with its superior resistance to intercept,resistance to decline and strong ability of multiple access network,it has been widely used in the military and civilian communications.It belongs to the typical non-stationary signal as the frequency of frequency hopping signal changing with time,it is hard to analysis the frequency hopping signal accurately using analysis method of the simple time domain or frequency domain,and the time-frequency analysis technology provides a very effective method to deal with this kind of non-stationary signal.The basic idea of the time-frequency analysis is to design the joint function of the time and frequency,using it describing the signal energy density or strength in different time and frequency at the same time.In recent years,the related research developing and deepen constantly,its application scope expands unceasingly,it become a very active and important research subject in the field of modern signal processing.In this thesis,according to the time-varying characteristic of the frequency hopping signal,the time-frequency analysis method suitable for frequency hopping signal is further studied,the main work includes:Firstly,the main performance index of frequency hopping communication system is in-depth studied along with the main line of time-frequency analysis,and a frequency hopping signal mathematical model and the standard frequency hopping pattern in the research are established combining the feature of frequency hopping signal in the time domain.Secondly,two kinds of time-frequency analysis method STFT(Short Time Fourier Transform)and SPWVD(Smoothed Pseudo Wigner-ville Distribution)are studied,and their own performances of the two different time-frequency analysis methods are compared.Thirdly,because it exists serious cross-term interference,large amount of calculation and poor time-frequency clustering problems in the time-frequency analysis method,the image segmentation and image template matching processing of frequency hopping signal detection method and morphological image processing offrequency hopping signal detection method are studied,and they are simulated with the MATLAB software.The distribution performance of two time-frequency analysis method STFT and SPWVD are improved largely after morphological processing,and the SPWVD algorithm has good clustering compared with STFT algorithm overall comparison,no matter distribution before through morphology processing,or after processing distribution.Based on signal detection rate of this algorithm analysis,it is found that under the premise of jump speed of the frequency hopping signals is constant the interference signal is effectively restrained using both the theory of morphology and the improved method of image processing,and a certain range of SNR signal detection is realized.The algorithm has certain feasibility and it is a kind of effective detection based on the time-frequency analysis method of frequency hopping signal. |
