Study On Sorting Technology Of Variable Speed Frequency Hopping Signal

The continuous development of communication technology has created conditions for the application of variable speed frequency hopping communication technology.The variable-speed frequency hopping communication technology adopts the strategy of“speed change” to pose a new challenge to communication against reconnaissance.The reconnaissance technology for variable frequency hopping signals mainly includes signal interception,detection and sorting.Compared with the fixed-speed frequency hopping signal,the "double-change" characteristic of the variable-speed hopping signal makes it difficult for the scouting party to sort the signal.The application of the variable speed frequency hopping communication system puts a severe test on the sorting of communication signals in complex environments.To this end,it is necessary to carry out research on its sorting technology for the characteristics of variable frequency hopping signals,and provide a practical and feasible signal sorting scheme.In this paper,the blind source separation idea is used to study the sorting technology of the noisy mixed signal composed of fixed frequency signal,fixed speed frequency hopping signal and variable frequency hopping signal.Main work of this paper is as follows:(1)Target signal constructionBased on the working principle of the communication system,the mathematical model of the fixed frequency signal,the conventional frequency hopping signal and the variable speed frequency hopping signal is established,and the mixed target signal model is constructed to generate the target data.Based on the generated simulation data,the actual collected data is analyzed.(2)Data preprocessingFor the problem that the received mixed signal contains noise and other devices interfere with the target signal,the mixed signal is preprocessed to increase the independence between the components of the mixed signal..The pre-processing methods used include de-average,whitening and estimation of the number of signal sources.In the case that the number of receiving antennas is less than the number of signal sources,combined with the characteristics that the variable frequency hopping signals are insensitive to higher order cumulants,a 4cum-GDE estimation source number algorithm based on fourth-order cumulants is proposed to improve the number of sources of signals.Estimated ability.Experiments are carried out to verify the effectiveness of the proposed algorithm.(3)Mixed signal separationBecause separation is the core of sorting,this paper uses blind source separation to study the separation signal of mixed signals.According to the relationship between the number of receiving antennas and the number of signal sources(positive and underdetermined),the corresponding algorithm is proposed to solve the problem of separation of variable frequency hopping signals.For positive definite mixing,the focus is on the separability of the variable speed frequency hopping signal in the mixed signal.On the basis of demonstrating that the variable speed frequency hopping signal can be blindly separated by independence,the separation of the mixed signal containing the variable speed frequency hopping signal is completed by using the independent component analysis algorithm.For the under-determined mixing,combined with the frequency-domain sparseness of the variable speed hopping signal,separation of mixed signals using sparse component analysis algorithmthe.At the same time,because the mixed matrix estimation algorithm for fixed-speed hopping signals generally needs to be combined with the estimation of the hopping time and the extraction of the single-hop signal,,the random change of the hopping speed of the variable speed hopping signal makes it difficult to estimate the hopping time.An improved "two-step method" algorithm is proposed.Firstly,the "adaptive noise threshold eigenvalue decomposition + density peak clustering" algorithm is proposed to estimate the hybrid matrix,and then a regularized adaptive matching algorithm is proposed to recover the source signal.The effectiveness of the two algorithms proposed in this paper is verified by experiments.(4)Mixed signal sortingFor the separated multi-frame signal,the blind source separation algorithm cannot judge whether the separated signals in different time periods are from the same source signal,it is impossible to judge whether the separated signals in different time periods are from the same source signal,and the "double-change" characteristic of the variable speed frequency hopping signal leads to the problem that it is difficult to estimate its parameters.A sparse Bayesian estimation algorithm and frame overlap are proposed.The overlap algorithm sorts the variable speed frequency hopping signals of different receiving time periods to complete the transition from separation to sorting.The effectiveness of the sorting algorithm is verified by experiments.