Research On The Method Of Coherent And Non-coherent Signal Processing Of Distributed Radar

With the development of detection accuracy and power of radar system,traditional phased array radars are required to have the characteristics of high power and large aperture.However,large-scale phased array radars are confronted with the problems of poor maneuverability and expensive cost.For this reason,the concept of distributed coherent radar was first proposed by Lincoln laboratory.The key technology of distributed coherent radar is to coordinate multiple small aperture radars,and obtain the detection performance of large aperture radar by signal level coherent processing.However,when distributed coherent radar works,the targets need to satisfy certain coherence conditions.Moreover,distributed coherent radar requires strigent space-time synchronization and phase synchronization which is difficult to achieve in real radar systems.It has been shown that with non-coherent processing,a target's RCS spatial variations can be exploited to obtain a diversity gain for target detection and parameter estimation.Therefore,the research of distributed non-coherent signal processing methods is of great significance for the further development of distributed radar.This thesis mainly studies distributed coherent/non-coherent signal processing methods.Firstly,the signal model of distributed coherent radar is established and the influence of the estimation error of coherent parameters on the coherent synthesis efficiency of distributed coherent radar is studied,while the boundary conditions of distributed coherent radar satisfying the coherent requirments are also researched.Secondly,the problems of clutter suppression in distributed non-coherent radar are surveyed.Among them,the clutter suppression technology in the case of self-receiving monostatic radars is relatively mature.This dissertation mainly discusses the problems of clutter suppression in the case of bistatic radars.To solve the problem of non-stationary distance of clutter and difficulty of suppression of bistatic radars,a clutter suppression algorithm based on bistatic DDMA-MIMO radar is proposed.Finally,the distributed non-coherent fusion detection and target parameter estimation are studied,the centralized detector and the distributed dual-threshold detector are deduced,and an estimation algorithm of target position and velocity based on matching matrix is proposed.The main contents of this thesis are given as follows:1.The distributed coherent radar signal model and the analysis of the coherent boundary conditions are researched.Firstly,the signal models of coherent receiving and coherent transmitting/receiving modes for distributed coherent radar are established.Secondly,the influence of the estimation error of coherent parameters on the efficiency of coherent synthesis is analyzed.Finally,the antenna size of distributed coherent radar is analyzed under three constraints: spatial correlation constrains,coherence parameter constraints and beam dwell time constraints.2.The bistatic clutter suppression is researched.Firstly,the clutter characteristics of bistatic phased array radars are studied.Moreover,the simulation results show that Doppler has the characteristics of range non-stationary in bistatic clutter.When STAP is applied to the received echoes,the performance of clutter suppression is seriously damaged.To solve this problem,the clutter suppression method based on bistatic DDMA-MIMO radar is proposed.Secondly,clutter characteristics of bistatic DDMA-MIMO radars are studied.The clutter is distributed in a plane of three-dimensional space.When the radial velocity of the target is non-zero,the target does not change with the distance,showing distance-independent.This characteristic is conducive to select IID samples and improving the clutter suppression performance.Finally,the validity of algorithm is verified by simulation analysis.3.The target detection and target parameter estimation are researched.Firstly,centralized detection and distributed dual-threshold non-coherent fusion detection method are analyzed.Secondly,to tackle the problem of high computation complexity in maximum likelihood estimation of target parameters,a target parameter estimation method based on matching matrix is proposed.The method uses clutter information to construct matching matrices for clutter correspondence of different receivers.According to the fact between target and clutter in the same location,the target parameters are estimated by matching matrices.Finally,the validity of algorithm is verified by simulation analysis.