Study On Time Delay Estimation Of Narrow-band Signals In Wireless Passive Position

This thesis studies the time delay estimation (TDE) under the situation of passive source positioning based on the received wireless signals. The research object is to find algorithms aimed at the TDE of narrow-band signals. With the rapid development of wireless technology, wireless positioning service is more and more bound up with people's daily life. For instance, LBS can provide people with various kinds of location information with the help of combining mobile network, wireless positioning and GIS. TDE plays an important role in passive wireless positioning, of which the precision and efficiency directly determines the performance of the entire positioning system. Thus, it is of great value to research into the topic of TDE not only in theory but also in application.Three main factors that deeply affect the TDE precision of narrow-band RF signals have been thoroughly analyzed; they are frequency difference (FD), bandwidth and SNR. Considering the above factors, a frequency difference solution and three TDE methods have been proposed in this thesis, which will be introduced briefly as follows:Firstly, the TDE model under the existence of frequency difference is derived and compared with the classical TDE model, obtaining the important influence of frequency difference to TDE precision. Thus a frequency difference compensation (FDC) solution and an improved frequency difference estimation method based on frequency domain correlation are proposed here. Theoretical analysis and simulation experiments show that this frequency estimation method can give satisfactory results for narrow-band signals under relatively low SNR.Secondly, considering the problem of bandwidth and SNR for narrow-band RF signals, a waveform comparison TDE method based on continuous wavelet transform (CWT) is proposed. This method at first proves the feasibility of combination between CWT and other TDE methods using the shift-invariance characteristic of CWT, thus enlarging the application scope of CWT in solving TDE problem. At the same time, this method is insensitive to bandwidth due to the adoption of waveform comparison (WFC) method. Simulation experiments verify that this method has a relatively high TDE precision.Thirdly, the phase spectrum TDE model under the existence of frequency difference is derived in this thesis. Considering also the factors of bandwidth and SNR, a phase spectrum TDE method based on FDC is proposed. In this method, CZT transform is applied to increase the available frequency bins within a limited bandwidth. Meanwhile, kalman filter is also adopted here because of its excellent anti-noise characteristics.Fourthly, this thesis theoretically proves that MUSIC method is applicable in solving TDE problem after comparing TDE model with DOA estimation model and harmonic frequency estimation model. Thus a MUSIC TDE method based on FDC is proposed, and the MUSIC TDE model under the existence of frequency difference is also derived. Simulation experiments and real date tests demonstrate the effectiveness of this method.Finally, the TDOA measuring system for narrow-band RF signals is improved, which is operated under the environment of Matlab. Real data have also been adopted to verify the ability of various kinds of TDE methods.