Researches On Frequency Hopping Communication Signal Detection And Parameter Estimation Method

Frequency Hopping technology has become a major technology to anti interference in communication at home and abroad. Nowadays, Conducting research on frequency hopping communication has been one of the important tasks in the field of communication countermeasure. To complete the combat mission of frequency hopping communication confrontation, it needs to intercept frequency hopping signal effectively. In the situation that enemy communication signal is unknown, the recevier needs to detect the received signal firstly, judging whether there is a frequency hopping signal. Under the premise of confirming that frequency hopping signal is present, the next step is to reconnaissant the activity patterns of that frequency hopping radio, that is to say, the receiver needs to eatimate the parameters of frequency hopping signals without any prior information. Correctly estimating the parameters of frequency hopping signal provides an important foundation to get access to enemy communication information and make effective interference. Contents of this paper can be divided into three components: frequency-hopping signal detection, parameter estimation and symbol rate estimation.In the part of frequency hopping signal detection, detection algorithm based on time frequency analysis has been improved. This part reserches a method of setting the threshold to eliminate the background noise based on energy statistics, solving the issue in the original algorithm that it requires to place the threshold multiply to observe de-noising effect and then determine the denoising threshold. In addition, this part introduct an adaptive algorithm based on SDM to set denoising threshold in the process of differential sequence of time-frequency ridge lines denoising, achieving the detection of frequency hopping signal under the complex electromagnetic environment.In the part of hopping signal parameters estimation, this paper introduces the idea of time-frequency filtering firstly, and makes the time frequency inverse transform for the time-frequency distribution which has been elimated the interference signals in the part of signal detection to obtain a clean frequency hopping signal, providing a more clean time-domain signal for the parameters estimation, de-hopping and demodulation of frequency hopping signals. Then this part reserches hopping parameters estimation method based on time frequency analysis, and get a algorithm to detection the interfered hop of signal in the situtation that frequency-fixed signals come into collision with frequency hopping signals. That algorithm can detect the hopping frequency, the start and end time of the interfered hop of siganl, achieving the blind parameters estimation of frequency hopping signal.In the part of frequency hopping siganl symbol rate eatimation, this paper introduces the symbol rate estimation method based on signal envelope spectrum to estimate the symbol rate of frequency hopping signals firstly.Then this paper analyzes the background noise produced by the envelope transform and the best tap length in the line enhancement process. Also the symbol rate estimation method based on the power spectrum of signal-square is introduced into the symbol rate estimateation of frequency hopping signals and has been improved. Both methods achieve to estimate symbol rate of the frequency hopping signals blindly, first modulation of which is linear.Then this part dehop the frequency hopping signals,first modulation of which are OQPSK and MSK, through the hopping parameters obtained in the part of parameters estimation. After that we estimate the symbol rate using the de-hop signals by the OQPSK and MSK signal symbol rate estimation methods respectively, achieving symbol rate estimation of the frequency hopping signals blindly, first modulation of which are OQPSK and MSK. While researching on the three parts, this paper has been doing some experiments to verify the validity and accuracy of the algorithms.The simulation results show that these methods are suitable for the actual project.