Adaptive Detection Of Target With MIMO Radar

With the increasing complexity of the working environment,the challenges faced by radar are becoming more and more grim.Therefore,it is very significant to study and explore better target detection algorithms.As a new radar system,multiple-input multiple-output(MIMO)radar has became a research hotspot in the field of radar in recent years because of its unique advantages.Unlike traditional phased array radar,MIMO radar transmitters emit independent or even orthogonal waveforms to obtain more comprehensive target information.Therefore,MIMO radar can obtain waveform diversity performance and improve target detection performance.In this thesis,the adaptive detection of MIMO radar is studied,and the detection algorithms are designed based on a prior knowledge of clutter.The main work and contributions of this thesis are as follows:1.The basic principle of MIMO radar is introduced and the signal model of MIMO radar is established.Then,the signal processing flow of MIMO radar is elaborated in detail.Statistical analysis is made on the measured radar data,including the amplitude and power spectrum characteristic.Then simulation experiments are carried out using the measured data.The results show that compared with the adaptive detection algorithms,the traditional signal processing method has more false alarms.2.Under the background of Gaussin clutter,the adaptive algorithms based on the structure of clutter covariance matrix are studied for target detection with MIMO radar.Based on the generalized Likelihood Ratio Test(GLRT)and Rao test,three adaptive detectors are designed by exploiting the persymmetric structure of clutter covariance matrix.In particular,training data are not required in the proposed adaptive detectors.Furthermore,analytical expressions for the probability of false alarm and detection probability are derived,which are confirmed by Monte Carlo simulations.The analytical expressions for the probability of false alarm show that the proposed detectors exhibit Constant False Alarm Ratio(CFAR)property against the clutter covariance matrix.Both the simulated data and measured data show that the performance of the proposed detectors are better than that of the traditional adaptive detectors.Among the three detectors,the P-AMF test is the most robust in the mismatched case.In addition,another contribution is to propose a two-stage detector whose first and second stage coincide with the P-AMF and P-Rao tests,respectively.By choosing different threshold pairs,the robustness or selectivity of the two-stage detector can be tuned.3.Under the background of Gaussin clutter,three adaptive algorithms based on Bayesian theory are studied for target detection with MIMO radar.The clutter covariance matrix is modeled to be unknown and stochastic,and the inverse complex Wishart distribution incorporating the prior knowledge is assigned to the covariance matrix.According to the criteria of the GLRT,Rao test and Wald test,three adaptive detectors in the Bayesian framework are proposed.The proposed detectors have two main advantages: 1)no training data are required;2)a prior knowledge about the clutter is incorporated in the decision rules so that the performance gain can be obtained.Numerical simulations show that the proposed Bayesian detectors outperform their non-Bayesain counterparts,especially in the case of small sample number of the transmitted waveforms.When a mismatch in the receive steering vector happens,the proposed Bayesain Wald detector has the best robustness,and the Bayesian Rao detector has the strongest rejection capability of mismatched signals.