Research On Some Problems In Signal Detection For MIMO Radar

MIMO technique has been widely employed in communication system to combatmulti-path effect and improve transmission rate.Inspired by the SIAR(SyntheticImpulse and Aperture Radar) and MIMO communication technique,two kinds ofMIMO radar are put forward recently.The first one is based on both transmitting andreceiving diversities that is firstly presented by Eran Fishler.It is referred to TR-MIMOradar(Transmitting Receiving MIMO radar) or S-MIMO(Statistical MIMO radar).Inconventional radar,detection performance will be impacted when target RCSfluctuating.However,in TR-MIMO radar,a target's RCS spatial variations can beexploited to improve system performance.TR-MIMO has simple hardware structureand big effective aperture.Furthermore,it can combat target RCS fluctuating throughspace diversity.The other kind of MIMO radar is based on only transmitting diversitythat is firstly put forward by D.J.Rabideau.It is referred to T-MIMO(TransmittingMIMO) which has the same antenna configuration with the phased array radar.Therefore,it can be considered to be an extension of phased array radar.On transmit,the elements(or subarrays) of T-MIMO radar transmits omnidirectional,orthogonallycoded waveforms.On receive,the DBF(Digital beamforming) is employed.It has beenshown that waveform diversity enables the MIMO radar superiority in severalfundamental aspects,such as significantly improved parameter identifiability andangular resolving power.In this dissertation,the target detection for these two kinds of MIMO radar isinvestigated,and the main research focus on the following issues:1.The principles and characteristics of TR-MIMO radar are studied.Basing onthe classic RCS model,x~2,the detection performance of TR-MIMO radar is analyzed.And then,the detection performance of TR-MIMO radar is compared with that ofphased array radar when the RCS modeling is Swerlingâ… .It is conclude that at lowsignal noise ratio(SNR),the MIMO system outperforms the phased-array system.2.A new detection algorithm for MIMO radar based on the Hidden Markovmodels(HMM) is proposed.The distance among antennas is large in TR-MIMO radar.So,the target is observed from different aspects.Therefore,the returns of targetexhibit anisotropic,which is opposite to the returns of clutter.According to thischaracteristic,the target can clutter can be modeled with different HMMs respectivelyto realize the separation of them.3.The principles and characteristics of T-MIMO radar are studied.Thecharacteristic and the signal model of the system are investigated firstly.A simulationplatform for T-MIMO radar is built,through which the MIMO radar performance and itsadvantages over the traditional phased radar are verified,such as anti-interception ofradar signal,weak target detection in strong clutter,and low velocity target detection.Furthermore,the detection performance is studied through signal noise ratio.It isconcluded that when the LFM signal is employed,the MIMO system has the sameperformance with conventional radar.4.A preliminary investigation into space-time-waveform adaptive for T-MIMOradar is presented.Existing space-time adaptive processing(STAP) algorithm isextended to the waveform-time-space adaptive processing(WTSAP) case.The signalmodel required to generate simulated data is developed.Simulation results show asignificant improvement in detector performance as compared to conventional phasedarray radar.5.The Hough Transform(HT) algorithm is applied to MIMO radar detection.Amodified algorithm is proposed to overcome the computational complexity.Theslope-intercept space is employed and the HT is realized through shifting the parameterspace cells.The computational load is reduced when processing the data with the samearrival time.Furthermore,the phase information of returns are exploited to realizecoherent integration,the detection performance can be improved when signal noise ratiois low.Finally,the method of using HT in long time integration is studied.