Researches On Frequency Hopping Signal Feature Extraction And Signal Sorting Method

Wireless communication is the main way of modern military operations,in order to make the communication can be safe and reliable,frequency hopping communication technology came into being.Frequency hopping communication technology attaches great importance by the military because of its superior characteristics,the parties are actively engaged in frequency-hopping communication technology research,and have made great progress.As the actual combat environment is more complex,so the intercepted signals often contains more than one frequency hopping signal.In order to complete the combat mission,we have to sort the intercepted mixed signal,so that we can learn useful information on a single frequency-hopping signal.However,the effect of signal sorting depends on the accuracy of frequency hopping signal parameters estimation and the performance of the signal sorting algorithm.It can be seen that the feature extraction and sorting of frequency hopping signals are two important aspects in the practical combat application.It is of great significance to carry out the research on these two aspects.This paper will focus on single antenna reception and array antenna reception to start research.In the part of characteristic parameters extraction,as there are usually a variety of interference signals and noise in the actual communication environment,so we first need to carry out pre-processing on the frequency hopping signal.Firstly,the time-frequency analysis of the intercepted signal is carried out,and then the gray signal is carried out,the interference signal and noise are eliminated by the morphological method,which provides a clean frequency hopping signal for the subsequent work.Then,we focus on the frequency hopping signal parameters estimation based on the quadratic difference algorithm and the SDIF algorithm under the condition of single antenna reception,and the DOA estimation based on the MUSIC algorithm in the case of array antenna reception.Finally,the accuracy and feasibility of the proposed algorithm are verified by simulation experiments.In the part of frequency hopping signal sorting,for different situations,this paper analyzes and studies different sorting algorithms.Firstly,the frequency hopping signal sorting based on the filtering method is studied,and this method is used to deal with the situation that there is no overlap between the frequency sets of different frequency hopping signals.Although the application of this method is limited,but in this particular case can be very good to achieve the signal sorting.Secondly,the algorithm of signal sorting based on time arrival method is studied.And this algorithm is computationally small and easy to implement real-time sorting.Then,the algorithm of frequency hopping signal sorting based on time-frequency domain information and time-frequency-spatial domain information is studied,and it utilizes the method of combining CDIF algorithm with CT sequence s earch to realize frequency hopping signal sorting.Finally,the method of frequency hopping signal sorting based on clustering is studied,and the advantages and disadvantages of several common clustering algorithms are analyzed and compared.The algorithm used in this paper is based on the traditional KHM clustering algorithm,and it has been improved,the accuracy and feasibility of the algorithm are verified by simulation.