Reseach On Technology Of Pulse Signal Deinterleaving

The task of pulse signal sorting system is to deinterleave the received pulses signal into groups,where pulses from the same emitter are assigned to the same set of pulses. Pulse signal deinterleaving is the base of identification of the emitter and the further signal processing. In modern complex electromagnetic environment, pulse density increase significantly and the pulse signal waveform become much more complicated. As the basis for the subsequent signal processing, effective and rapid pulse signal sorting become a key step of electronic surveillance systems. So that scholars emphasis on the research and technology of pulse signal deinterleaving, and achieved great results.Conventional pulse signal deinterleaving depends on each pulse parameters as the separation characteristic, i.e. the PDW(pulse description word,including pulse arrival time, the arrival direction of the pulse, the pulse width, the pulse carrier frequency and the pulse amplitude). First lower the pulse density by using the fixed pretreatment tolerance in the pretreatment.Then use pulse arrival time(TOA) parameter of the pulse burst to estimated pulse repetition interval(PRI) and deinterleave pulse sequence. Traditional sorting methods are simple and easy to implement, which can effectively sorting signal of conventional system, so that they are widely used in practical engineering applications. In the current complex signal environment, pulse parameters are changeable and the parameter space overlap seriously. Traditional sorting method for sorting non-cooperative complex intercepted pulse signal get deteriorated effect. We urgent need for more information for the pulse deinterleaving. This thesis focuses on the non-cooperative complex pulse signal deinterleaving technology research and practice, which main contents are as follows :1. In the analysis of the current complex signal types, based on the summary of the shortcomings of traditional pulse sorting system structure and proposed for complex signal environment of the new structural framework sorting, sorting and combining a new framework for inter- pulse parameters to start sorting, pulse the sorting feature extraction and clustering technology research in three areas.2. Based on analysis of the three kinds of classic sorting methods- CDIF,SDIF and PRI transform in-depth analysis, simulation results show that the pulse parameters deinterleaving method can achieve a fixed PRI pulse signal sorting. Based on the PRI transformation method, an improved method in order to adapt the jitter pulses signal sorting was put forward, and simulation shown that improved method for the jitter pulse of complex signals environment had a good effect.3. Intra-pulse modulation characteristics was analysed by using the wavelet transform and fuzzy function. Pulse features with good separability were extracted for sorting. Wavelet ridge extraction algorithm is improved, and finally the simulation proved wavelet ridge characteristics and ambiguity function section features had a good anti-noise and clustering performance, which can effectively achieve the pulse signal pulse modulation in different sorting.4. Clustering sorting methods Advanced FCM clustering algorithm was researched and compared their advantages with traditional C-means clustering algorithm. Based on summing up the performance of FCM algorithm, analysis of SAKM algorithm is introduced. Through the simulation,we find that the SAKM algorithm which has better adaptability is good at real-time sorting and more suitable for unknown emitter signals.