Emitter Individual Identification Based On Parameters And The Transform Domain

Individual characteristics naturally attach to the signal due to the individualdifferences of the signal transmitting apparatus and they can be used to identifydifferent transmitting apparatus. The main job of individual identification is to extractthe individual features from the intercepted signal and use these features inclassification and recognition. Individual identification is not only a necessary mean ofelectronic warfare, is also an important mean of communication equipment,authentication, identification and other applications. This paper aims at researching thisrapidly developmental issue and the main study as follows:Firstly, this paper investigates the formation mechanism of the individualcharacteristics and establishes the signal model with additional individual characteristics.Then the paper analysis the transient-state features and the steady-state feature, and getthe way to achieve transient-state signal and steady-state signal. A process of theindividual characteristics abstraction is proposed. This process gives two researchdirections on individual identification. One is based on parameters and one is based onconversion domain processing.Secondly, the individual identification based on parameters is discussed. Throughthe research of the chaos theory and the fractal theory, we can think these theories haveimportance effects on signal with strong nonlinear characteristics. We find Lyapunovparameter and step dimension can jointly identify different signal effectively. And thenthe algorithm of individual identification based on parameters is proposed and is testedin the followed simulation experiment.Finally, the individual identification based on conversion domain processing isdiscussed. The conversion domains which include abundant individual characteristicinformation are researched, such as ambiguity function domain and wavelet packetdomain. And the paper research the algorithm used to reduce dimension of thesedomains. The improved rapid algorithm based on ambiguity function domain isproposed. During the algorithm research, the negative influence of modulationinformation is discovered. And we find that the high order domain process can inhibit the modulation information of MSK, BPSK, and QPSK signal. Then the algorithmbased on wavelet packet domain faced steady-state signal of MSK, BPSK, and QPSK isproposed. And the simulation experiment is followed to test the performance ofalgorithm.