Research And Realization On STFT-based Detection Technology Of Frequency Hopping Signals

The emergence of the frequency hopping(FH)communication brings a new subject to the field of signal detection.For the detection of non-cooperative FH signals,domestic and foreign experts and scholars have proposed various detection theories.But in the field of engineering,the practical application results of detection of FH signals are less,and they basically used Field Programmable Gate Array(FPGA)to develop related systems.In this thesis,the computational complexity of the current detection algorithm of FH signals is analyzed,and a time-frequency analysis method that is more conducive to hardware implementation is given.In view of the shortcomings of FPGA's scarcity of storage resources and poor data processing flexibility,a scheme of implementing a detection system of FH signals on the Zynq platform is given.The main contents of this article are as follows:1)This thesis firstly introduces the basic principle of FH communication,analyzes the influence of signal modulation method on the frequency spectrum,and expounds the reason for choosing Minimum-Shift Keying(MSK)or Continuous Phase Frequency Shift Keying(CPFSK)as the modulation method for FH signals transmission.Secondly,this thesis introduces two types of detection methods of FH signals,energy detection and time-frequency analysis detection,and compares the detection effect and computational complexity of each common time-frequency analysis method.Finally,this thesis chooses the short-time Fourier transform(STFT)as the time-frequency detection method.2)This thesis makes an analysis of STFT which is the most classic and basic method,and points out that the traditional algorithm based on fast Fourier transform has a huge computational burden.In order to reduce the amount of calculation and facilitate hardware implementation,this thesis presents a time-frequency formula that is more conducive to the implementation of the algorithm by increasing the step size of the STFT sliding window.At the same time,in view of the contradiction between time and frequency resolution of STFT,this thesis uses the principle of star sensor to locate stars,and uses the centroid method with threshold to improve the frequency detection accuracy in the case of short windows.3)In this thesis,the complete detection process of FH signals is given,and the specific time-frequency distribution data processing process and method are analyzed and simulated in detail.Based on the idea of building a soft core in FPGA,this thesis applies the STFT-based detection and identification technology of FH signals to the Zynq platform by using the method of FPGA+ARM hard core.At the same time,this thesis analyzes in detail some details that need to be paid attention to in the system design process,and makes a simple evaluation of the final system performance,which verifies the achievability of the algorithm and the advantages of the Zynq platform.