Detection And Parameter Estimation Technology Of Frequency Hopping Signals Based On Time-frequency Analysis

Frequency hopping(FH)communication is a common method of spread spectrum in spread spectrum communication systems.Because of its advantages such as low interception probability,excellent confidentiality and anti-interference,it is used in military reconnaissance and anti-interference field,and in addition,it plays an important role in civil communication systems.Therefore,in-depth research on frequency hopping signal detection and parameter estimation techniques is very important.The main content of this paper includes time-frequency(TF)analysis and parameter estimation of FH signal in a non-cooperative environment.This paper mainly studies the TF analysis of FH signal based on the extended modified B-distribution algorithm,the parameter estimation of FH signal using TF moments and their improvements.Firstly,this paper introduces the generation mechanism of the FH signal and gives a mathematical model.It briefly explains the technical specifications of the FH communication and the parameters of FH signal to be estimated later.The commonly used TF analysis is studied from both linear and bilinear aspects.The analysis performance of different algorithms is compared through simulation,and the advantages and disadvantages are summarized.In addition,this paper studies two basic parameter estimation algorithms of FH signal.Next,this paper focuses on a new TF analysis method based on EMBD.The algorithm has a significant effect in removing cross-term interference.The signal is mapped into the ambiguity domain,and the selected Doppler-lag kernel function is used for two-dimensional low-pass filtering.The processed signal in TF domain can be obtained by two Fourier transforms.Compared with the basic TF analysis method,the TF analysis algorithm based on EMBD is more capable of suppressing noise.The TF diagram is more "clear" under the same signal to noise ratio(SNR),and the parameter estimation of FH signal is more accurate.However,the real-time performance of the algorithm is not high.In view of this limitation,this paper improves the kernel function involved in the algorithm and replaces the hyperbolic window with a Gaussian window.Simulation results show that the TF analysis effect of the improved algorithm under high SNR is close to that of the previous algorithm,but the running time is shorter,and the relative errors of the estimated hopping time and duration are smaller.Finally,in order to further improve the accuracy of the parameter estimation of FH signal,starting from the TF diagram modification,an energy detection algorithm based on adaptive threshold is studied.The threshold of this algorithm is determined by the energy distribution of the data in each segment of the TF matrix.The signal segment and the noise segment are divided according to the relationship between the short-time energy and the threshold.At the same time,this paper improves the parameter estimation of FH signal using TF moments.After selecting the threshold to process the TF moment,the influence of noise is suppressed,thereby improving the accuracy of the algorithm in estimating the parameters of FH signal.Combined the TF diagram modification algorithm with the improved parameter estimation of FH signal using TF moments,the parameter estimation error is further reduced,and the simulation of MATLAB is used to prove the feasibility of combining the two algorithms.