Research On The Key Techniques Of Signal Processing For Spaceborne Early Warning Radar

Spaceborne early warning radar can be used to implement ground moving target indication(GMTI)or air moving target indication(AMTI).However,the clutter Doppler spectra of spaceborne radar extend severely due to the high moving velocity of the satellite platform.Additionally,severe non-homongeneous clutters caused by broad illuminating areas also limit the performance of clutter suppression.In contrast with the clutter suppression and target detection in airborne radar,it is more challengeable for spaceborne radar.Hence,it is worthy of further researching the key signal processing methods for clutter suppression and MTI in spaceborne early warning radar.In this thesis,through the aspects of radar system and space-time adaptive processing(STAP),we emphasize on those studies for the STAP method based on Phased-MIMO radar,the STAP technique based on higher order statistics(HOS),the one-dataset STAP approach based on sparse recovery(SR)technique and the STAP method based on nested array and nested pulse repetition interval(PRI).This thesis first introduces the research status domestic and overseas with respect to the Spaceborne early warning radar system.Particularly,it stresses the research progress of clutter suppression and MTI.In addition,Phased-MIMO radar system is also discussed in chapter one.The second chapter mainly provides the model of the Phased-MIMO radar.According to the MIMO radar model,the signal model of the Phased-MIMO radar,the models of adaptive beamforming and non-adaptive beamforming are given.Next,this chapter carrys out simulations and analysis to illustrate that the Phased-MIMO radar is able to combine the advantage of the coherent processing in phased array radar and the advantage of the waveform diversity in MIMO radar,which is a tradeoff between phased array radar and MIMO radar.The third chapter studies the STAP technique based on Phased-MIMO radar.STAP model of the Phased-MIMO radar is offerred.Subsequently,is analyzes the STAP performances of filter output response,output signal-to-interference-plus-noise ratio(SINR)and minimum detectable velocity(MDV)via simulations.Finally,a novel array partitioning scheme based on STAP SINR is proposed.The chapter four studies the STAP method based on HOS in the case of phased array radar with a ULA.It proposes a novel SP-STAP method based on signal powers and provides the derivative process of the signal model of the proposed SP-STAP method.For contrast,the performances of filter output response,output SINR,output SINR loss,MDV and ICM effection for both the proposed SP-STAP method and the traditional STAP method are discussed through simulations.The fifth chapter studies the problem of target detection in the scence of non-homogeneous clutter and dense target.For the lack of independent and identically distributed(IID)training samples,this chapter proposes a novel deterministic-aided one-dataset STAP approach based on SR,which is referred to as DA-SDS-SRSTAP.This approach can realize clutter suppression without training sample but the usage of cell under test(CUT)data.It proposes another deterministic-aided generalized inner product(DA-GIP)algorithm to complete automatic recognition of the clutter components in the angle-Doppler power spectrum.This chapter describes the concepts of SR and space-time power spectrum firstly.Then it elaborates the signal model and flow of the proposed DA-SDS-SRSTAP method.At last,simulations have verified the effectiveness of the proposed algorithm.The sixth chapter studies the STAP method based on nested array and nested PRI.A perspective of difference co-pulse is proposed.Additionally,a novel STAP method based on optimum two-level nested array and optimum two-level nested PRI is also proposed in this chapter.According to the concept of difference co-array,the proposed method constructs a virtual space-time snapshot and a virtual clutter covariance matrix(CCM)through mathematic transformation of the original CCM.STAP filter operates on these virtual data.The proposed method can enhance system degrees of freedom(DOFs)and then improve STAP performance by mean of signal processing but not increasing the number of array element and transmitting pulse.