Direct Wave Extraction Technology In Passive Location System

In passive location system, how to attain an accurate radiant source signal is an important issue, it will decide whether the object could be located or not. In time-varying multi-channel communication system, the signal from radiant source then go through reflector and finally achieve to the receiver was called direct wave. The problem that how to extract the direct wave equals to the problem how to equalize the radiant source signal. Commonly, equalize technology was used to solve the problem.This thesis focuses on the time-varying multi-channel and amplitude signal to research the channel identification and signal equalization technology. The research consist several aspects as follows:First, the transmit properties of the time-varying multi-channel was introduced, include the transmit model and properties of the signal. The math model based on the channel transmit properties was offered in the paper at the same time; also, some kinds of channel model was described; include the generate method and application scheme, also the model simulation in Matlab;Secondly, the properties of the AM modulation signal was analyzed, including the modulation, demodulation principle and the meaning of other related parameters, more importantly, a special properties named cyclostationarity was testified which the algorithms depend on.Equalization technology is the mainly research content in the paper. Several blind equalization algorithms was summarized in the paper, and the blind algorithm based on second order statistics(SOS) was finally chose through comparison the performance and application background among different algorithms. Then, the SOS algorithm was detailed researched as a main algorithm. To be exactly, it consists of the following aspects: expounded the theory foundation, principle of the algorithm, the step to apply the algorithm and the performance; Analyzed the shortcoming of the algorithm, and given an improved approach to solve the problem. Furthermore, some new evaluation methods were applied, they are: correlation function and channel square-root error were used to evaluate the AM signal algorithm; channel square-root error and error vector amplitude were used to evaluate the 16QAM signal. Simulation based on Matlab indicated that the improved algorithm performed much better than the traditional one.This paper was supported by national space foundation.(NO:20060307)...