Research On Time Delay Estimation In Passive TDOA Location

With the rapid development of wireless communication technology, the wireless location technology is improving fast, the significant role it is playing in national defense construction and our daily life is becoming increasingly obvious. This thesis studies the time delay estimation in passive TDOA location. In the wireless localizing technologies, passive TDOA location has attracted more and more attention with the advantages of strong concealment, good flexibility and high precision. The time delay estimation technology is one of the key techniques in passive TDOA location, and its precision and efficiency directly affect the performance of the locating system. Therefore, making in-depth research of the time delay estimation technology has great significance in both theory and project application.This thesis systematically studies the existed time delay estimation theories and methods. Based on the analysis of their principles, the advantages and disadvantages, two improved schemes of the time delay estimation algorithms are presented. The organization is as follows:Firstly, the background and meaning of the project are summarized, and the characteristic, classification and application of passive location technology are concluded. Besides, the basic conception, locating model and performance evaluation index of the time delay estimation technology are briefly introduced. Further, the research situation and the development trend of the time delay estimation technology are analyzed. Moreover, the theoretical deduction of mathematical knowledge that is mainly used in this thesis is given, including the conception of higher-order statistics and Hilbert transformation, and some of their important properties are emphasized.Secondly, the GCCHTS algorithm is thoroughly studied. What’s more, according to the problem of low metering accuracy of this algorithm in correlated noise and low SNR, an improved scheme is presented, and an algorithm called TOC-CHTS is proposed. And then, the theory of TOC-CHTS is described in detail, and the time delay estimation step of TOC-CHTS is concluded as well. In TOC-CHTS, the signals are processed in advance. To put it in detail, the third-order cumulant of the homologous signals with noise are calculated first, and the correlation and Hilbert subtraction operation is implemented later. Both the theoretical analysis and the simulation show that the improved algorithm not only can depress the influence of correlated noise on time delay estimation, but also can attain time delay value precisely in low SNR. Above all, compared with the GCCHTS algorithm, the performance of the TOC-CHTS algorithm is much better.Thirdly, a time delay estimation algorithm called TOC-CATDE is studied in depth. In TOC-CATDE, since the time delay estimation attained is always biased in actual application, and it can’t work under the condition of variable SNR. Consequently, an improved approach is presented in this thesis, and an algorithm called MTOC-CATDE is proposed. In MTOC-CATDE, by adding a plus iterative factor, the tracing of variable SNR can be achieved. The principle and convergence property of MTOC-CATDE are systematically analyzed in this thesis, and the time delay estimation step of MTOC-CATDE is summarized as well. Both the theoretical analysis and the simulation indicate that MTOC-CATDE can attain the unbiased time delay value directly and accurately no matter the noise is uncorrelated, correlated, or the SNR is variable. And compared with TOC-CATDE, MTOC-CATDE has a relatively high time delay estimation precision. Over all, the performance of MTOC-CATDE is superior to that of TOC-CATDE.Finally, a summary of the overall research is made, and the future work on the subject is viewed.