Research On Parameter Estimation And Recognition Methods Of Step Frequency Radar Signal

The step frequency pulse signal has the advantages of small instantaneous bandwidth,small hardware requirements,and large synthetic bandwidth.It is successfully applied to modern radar systems.The characteristics of its large bandwidth and multi-pulse combination also bring certain difficulties to electronic reconnaissance.Complete the step frequency pulse signal reception and detection,identify the step frequency signal type and obtain its parameters,can provide intelligence support for military operations.In view of the current research on step frequency pulse signal reconnaissance,this thesis will focus on the mature single carrier frequency modulation step frequency signal,chirp step frequency signal,phase code step frequency signal and sinusoidal FM step.Study the characteristics of the four types of signals,study how to identify the four types of signals,and how to estimate the parameters of the four types of signals.The specific research content is as follows:1.Investigate the signal model of single carrier frequency modulation step frequency signal,linear frequency modulation step frequency signal,phase coded step frequency signal and sinusoidal frequency step frequency signal,and study its time-frequency characteristics and main description parameters.Research on two methods based on digital channelization receiving system and the combination of panoramic monitoring and digital filter bank receiving system to complete the step-frequency radar signal reception.This thesis studies the signal detection algorithms of step-frequency signals.It mainly studies the time-frequency domain energy detection,auto-correlation detection and wavelet denoising detection.2.Investigate the modulation type identification method of step frequency signal.The four types of step frequency signals use different modulation types,including single carrier frequency,phase encoding,linear frequency modulation,and sinusoidal frequency modulation.Based on the time-frequency characteristics of the signal,this thesis uses the phase difference principle and self-correlation theory to identify these four modulation types.After the modulation type is identified,the intra-pulse parameters are estimated according to each modulation type,where the carrier frequency is the main estimation parameter,and a high-precision estimation of the carrier frequency is performed using a correction algorithm based on the FFT frequency measurement.A fast demodulation algorithm is used to estimate the starting frequency and modulation parameters of the chirp signal.The auto-correlation algorithm is used to estimate the code rate of the phase encoding.And use the Carson criterion based sinusoidal FM signal parameter estimation algorithm to complete the parameter estimation of the sinusoidal FM signal.The validity and robustness of the algorithm are verified by experimental simulation data.Study the relationship between carrier frequency and arrival time to identify step frequency signals and other non-step frequency signals.3.Investigate how to correctly sort the signals of each radar in a multi-signal environment.In a multi-signal environment,only if the pulse signal of a radar is correctly sorted,can the radar be further discriminated whether or not the radar uses a step frequency type.This thesis studies the clustering algorithm as a radar sorting algorithm,and focuses on the hierarchical clustering algorithm based on the data field as a dense pulse stream sorting algorithm that includes the type of step frequency signals to achieve the step frequency pulse signal extraction.