Estimation Of The Time Varying TDOA And FDOA In GEO-LEO Dual-satellites Location System

In modern communication systems,satellite communication is a very important communication method,which can guarantee communication needs in extreme environments and emergencies.But satellite communication needs passive localization of interference source to ensure its system work from malicious interference.Fast and accurate passive localization of the interference source will directly improve the reliability of satellite communication.In this paper,the existing satellite-carried passive localization system which are wildly used,and the parameters TDOA and FDOA in the passive localization system are studied.Furthermore,the reasons for the large error of TDOA/ FDOA estimation and its elimination method are analyzed.This paper has studied the following aspects:1.Research on high-precision FDOA estimation method based on the fourth-order cumulant.In the existing TDOA/FDOA joint estimation methods,cross ambiguity function method and cross ambiguity function based on the fourth-order cumulant are usually used.In both of these methods,the fast Fourier transform method with high computational efficiency is often used to replace the correlation operation in cross ambiguity function,but the resolution of the FDOA estimation is reduced by the influence of the fast Fourier transform length.In this paper,we propose a high-precision FDOA estimation method based on the fourth-order cumulant and cross ambiguity function.Under the condition that the Fourier transform length is invariant,it is possible to improve the estimation accuracy of the FDOA only by adding a small amount of computation.2.Research on fast time-varying TDOA/FDOA estimation method.In the existing satellite-carried TDOA/FDOA estimation models,the narrowband signal is often used as the reference model,and the compression or broadening of the signal envelope caused by the Doppler effect is neglected.It is assumed that the TDOA/FDOA of the signal are mutually independent.However,the Doppler effect not only causes the FDOA,but also has an influence on the envelope of the signal,which reduces the direct correlation between the signals which are received by two satellite,resulting in an increase in the error of TDOA/FDOA estimation.In this paper,the signal model of TDOA/FDOA is changed with time.Firstly,the influence of TDOA/FDOA is eliminated by using the idea of motion parameter compensation.Although the performance of this method is very effective,the idea of using compensation requires a lot of computation time,which makes the algorithm less real-time and is not conducive to practical application.On this basis,this paper presents a fast time-varying TDOA/FDOA estimation method.For the TDOA/FDOA time-varying process,the FDOA is linearly changed,assuming that each FDOA estimate is approximately equal to the true value of the FDOA at the middle of the sampling data.The rate of change of the frequency difference is estimated by using the multiple sampling data.So it is possible to reduce the estimated errors and improve accuracy of the TDOA/FDOA estimation.The MATLAB simulation results show that the method proposed in this paper can greatly improve the TDOA / FDOA estimation accuracy approximate to the motion parameter compensation method and reduce the computation time.3.Research on the method of continuous fast time-varying TDOA/FDOA estimation.For the practical application of continuous passive positioning needs,and based on the fast time-varying TDOA/FDOA estimation method,this paper presents a continuous fast time-varying TDOA/FDOA estimation method based on least squares method.By linearly fitting TDOA/FDOA estimations of multi-segment continuous sampling data by least squares method,we can get the TDOA/FDOA of each time.It is verified in MATLAB simulation that this method can have smaller TDOA/FDOA errors than the fast time-varying TDOA/FDOA estimation method with multi-segment continuous sampling data.