| With the continuous progress and development of technical research in the communication field,the requirements for modern communications are getting higher.Frequency hopping communication technology has become the leading technology in the field of communication anti-jamming due to its low interception and high anti-jamming capability,and it has been applied to all walks of life.In electronic warfare,it is equally important to decipher the intercepted signal,in addition to ensuring the integrity and security of its own communication information.In this research background,it is studied in this paper that the parameter estimation of non-uniform frequency hopping signals in the environment with fixed frequency,sweep frequency,random burst interference and multiple noises.The key work content is divided into the following three parts:1.Preprocessing of frequency hopping signal time-frequency matrix.Based on the research of existing time-frequency analysis methods and the requirement of simulation accuracy,combined with histogram statistical method of image processing and adaptive threshold denoising method of signal processing,time-frequency matrix histogram denoising method,which counts time-frequency matrix data and selects thresholds by histogram is proposed in this paper.The simulation results show that the method is not affected by noise type,and can effectively distinguish noise points from effective signal points in the time-frequency matrix in the absence of noise information which meets the condition of blind estimation of signal parameters.Under the simulation conditions of this paper,the algorithm can achieve higher noise point removal rate and signal point retention rate when the signal-to-noise ratio is greater than-5dB.2.Parameter extraction of frequency-hopping signals without silent time.Aiming at the parameter extraction problem of no silent time frequency hopping signal,according to its time-frequency domain characteristics,a signal parameter set extraction method based on connected region labeling combined with time-frequency matrix time and energy characteristics of frequency hopping signals is proposed in this paper.The method extracts the time-frequency matrix information by using the connected region labeling method in image processing,filters and classifies the signal segments according to the time and energy characteristics of frequency-hopping signals,and finally selects the target signal set by comparing the signal-free duration.The simulation results show that the proposed method can have higher parameter identification capability under the condition of higher signal-to-noise ratio.3.Parameter extraction of frequency-hopping signal with silent time.Aiming at the problem of parameter extraction of non-uniform frequency hopping signals with silence time,this paper uses the mathematical morphology-based interference filtering method to remove most of the interference in the time-frequency matrix according to the different morphological features of the interference signal in the time-frequency diagram.The parameter set of the target frequency-hopping signal is selected based on the energy similarity of the homologous frequency hopping signal.The simulation results show that the proposed method has a good parameter recognition rate in the condition that the signal-to-noise ratio is higher than-4dB. |
