Radiation signal sorting method based on PDW multi-feature fusion
In view of the difficulty to deal with aliasing signals and low accuracy of sorting in the field of radiation signals sorting,this paper proposes a multi-feature fusion method,focusing on pulse description word(PDW),which is a two-step process architecture based on divide-conquer method.In the first step,the problem of radial source signal aliasing in the time-frequency domain is solved by the proposed spatial-time density clustering model.The second step is to further improve the accuracy of sorting through the intersection over union method.Compared with the traditional methods,this method can effectively separate the pulse signal of the mixed time-frequency domain with higher sorting accuracy.It is robust against the variation of pulse repetition interval(PRI)of radiation source.
radiation signal sortingspatial-temporal density clusteringintersection over union(IoU)multi-feature fusionpulse description word(PDW)