The Parameter Estimation And The Modulation Mode Recognition For Frequency Hopping Signals

Frequency hopping communications have also been well-documented merits for applied in military communications in 70 s because their excellent anti-jamming, anti-interception and the capabilities of multi-site networking. And they have also been applied for civil communications in recent years. The Frequency-hopping signals’ reconnaissance has become an important content for communication reconnaissance and the radio regulatory. The parameter estimation of Frequency-hopping Signals and the modulation mode recognition of Frequency-hopping Signals, two key technologies in Frequency hopping communications reconnaissance, have become the research hotspot of the Frequency hopping communications technology. Especially the estimation result and the recognition rate of Frequency hopping signals will determine the signals’ demodulated and information restore results. In this thesis, two novel algorithms have been proposed to complete the parameter estimation based on time-frequency analysis and Sparse Liner Regression techniques. they are the novel algorithm based on FFT and ALM, the novel algorithm based on STFT and ALM. Simulation results show that the two combined methods have higher estimation accuracy and better real-time. At the same time, relying on the analysis of fast frequency hopping communication system characters, a statistical algorithms has been proposed to resolve the recognition between FFH-BPSK signals and FFH-BFSK signals.The basic principle and the main technical indicators of Frequency-hopping communications are analyzed in the thesis, And the Frequency-hopping signals are simulated for subsequent parameter estimation and modulation recognition of frequency hopping signals in the end.Based on the applications of time-frequency analysis technologies in the parameter estimation of Frequency-hopping signals, This thesis presents the simulation results and performances analysis of parameter estimation of time-frequency algorithms such as bilinear time-frequency analysis(Wigner-Ville Distribution, pseudo Wigner-Ville Distribution and smooth pseudo Wigner-Ville Distribution), the short-time Fourier transform, Gabor transform, Hilbert-Huang transform and Augmented Lagrange’s theorem algorithm. The problem of low parameter estimations accuracy under low SNR channel is resolved combined with sparse linear regression techniques. At the same time, the short-time Fourier transforms and sliding window fast Fourier transform have short processing time, the hopping period estimation methods are further studied combining the high precision of the sparse liner regression and the good real-time of the STFT,Two new joint algorithm is proposed. Simulation results show that the method has better real-time and higher estimation precision. The modulation recognition methods are studied for FFH-BPSK modulation and FFH-BFSK modulation fast-frequency-hopping system. Finally, a better performance identification method is summarized by analyzing the characteristics of frequency-hopping communication system.