Study On Time-Difference Location And Its Related Issues Of Over-the-Horizon Passive Multistatic Radar

On the basis of both over-the-horizon radar and passive radar, over-the-horizon passive multistatic radar is a new type of radar developed to detect and locate sea-surface targets beyond visual range. This radar not only overcomes the shortcoming that the radar range performance of passive radars exploiting line-of-sight propagation is constrained by visual range, but also surmounts the drawback that the target resolution cell in conventional ground-wave over-the-horizon radars is quite large. Besides having good concealment and anti-stealth capability, it can provide both detection and high-accuracy location for over-the-horizon sea-surface targets. Therefore, over-the-horizon passive multistatic radar serves as an important approach to covertly obtain positional information of over-the-horizon sea-surface targets with a high degree of precision. Compared with conventional passive radar and over-the-horizon radar, it is more advantageous and has important research value and significance.On the research background of an over-the-horizon passive multistatic radar consisting of three transmitting stations and one receiving station, this dissertation concentrates on meeting the actual demand that target location accuracy needs to be improved for overthe-horizon sea-surface targets and addresses some key issues on time-difference location and its related issues for over-the-horizon passive multistatic radar deeply and systematically. The research mainly includes location for over-the-horizon sea-surface target based on time differences between target echo signals and direct wave signals with a multiple transmitters single receiver configuration, target location accuracy analysis, super-resolution TOA estimation for small bandwidth signals and superresolution DOA estimation for correlated signals. Meanwhile, the three-dimensional aerial target passive location problem is discussed in detail and a high-precision unambiguous aerial target location method based on time difference of arrivals of four receivers is studied in a single transmitter multiple receivers system. Theses researches enrich the framework of target location for passive radar. The main contributions of this dissertation can be briefly summarized as follows.In the first part, a passive over-the-horizon sea-surface target location method utilizing TOAs of target echo signals is studied. Firstly, the over-the-horizon passive multistaticradar system with three transmitting stations and one receiving station and the signal model are introduced. Then, the target location principle using three target TOAs which correspond to three transmitting stations respectively is elaborately explained. Next, a fast iterative algorithm to determine the target location is researched using spherical trigonometry. After that, the method of proper initial value selection is investigated. Finally, for the problem of target location ambiguity caused by the nonlinearity of the model, a hypothesis-testing based disambiguation algorithm is presented and it can solve the target location ambiguity problem successfully.In the second part, the localization precision analysis for the passive over-the-horizon seasurface target location method utilizing TOAs of target echo signals is studied. First, measurement precisions and performance evaluation parameters are given. Secondly, the principal factors that affect target location performance are analyzed. Moreover, it is pointed out that TOA measurement error and the geometric relationship among the transmitters, the receiver and the target are two major factors impacting the target location performance. Furthermore, the CRLB for target location is formulated and is used to analyze the optimal target location performance that can be attained theoretically. Lastly, the GDOP for target location is derived and is exploited to analyze the influence of geometric configuration among the transmitters, the receiver and the target on target location precision.In the third part, a super-resolution narrowband signal TOA estimation method based on compressive sensing is proposed. Firstly, the TOA estimation signal model using the sparse characteristic of narrow band signals is introduced. Secondly, in order to solve the problem that the TOA resolution using pulse compression is very low for narrowband signal, a novel TOA estimation method based on compressive sensing is proposed and it can achieve super-resolution TOA estimation for narrowband signal. Finally, to overcome the deficiency of large computational complexity and memory consumption in the traditional sparse reconstruction algorithms, an improved split augmented Lagrangian shrinkage algorithm with low computational complexity and memory occupation is presented and it improves the real-time performance of the TOA estimation.In the fourth part, a high-resolution DOA estimation method for correlated signals using sparse Bayesian learning is studied. First, the data association principle of utilizing DOAinformation of multiple targets to associate the TOA measurements with their corresponding target is expounded. Then, the sparse Bayesian learning theory of a multiple real-valued snapshots measurement model under maximum posterior probability criterion is introduced. Next, unitary transformation is exploited to transform the received data matrix of a uniform linear array into a real-valued extended data matrix and then the real-valued extended data matrix is sparsely represented under a real-valued sparse dictionary matrix. Therefore, the sparse signal model of DOA estimation based on unitary transformation is obtained. After that, an improved unitary sparse Bayesian learning method of DOA estimation is proposed. The proposed method is low in computational amount and good in angular resolution and can effectively estimate the DOAs of small angular separated correlated signals. Finally, the computational complexity of the presented method is discussed in detail.In the fifth part, a three-dimensional passive target location method using TDOA for aerial target is studied. Firstly, the system configuration and target location principle are described. An external illuminator serves as its transmitter. The TDOAs of the target measured by four receivers are exploited to locate the aerial target in three dimension. Then, the geometric relationship among the aerial target and four receivers is presented. Next, two scenarios in TDOA based aerial target passive location problem is illustrated: the first case is three constant TDOA surfaces with one intersection and the second case is three constant TDOA surfaces with two intersection. After that, an unambiguous target location solving algorithm based on target TDOA is proposed and the method of selecting a rational initial value of iteration is given. By using the above method, high-resolution unambiguous passive location performance can be achieved for aerial target. Then, the convergence of the proposed method is proved and the reason why the proposed method has fast convergence is figured out. Finally, the derivation of CRLB and GDOP for the three-dimensional target location problem based on TDOA is given.Theoretical and simulation results demonstrate that this dissertation has laid a firm theoretical foundation for engineering implementation of the system.