Study On The Theory And Method For Electromagnetic Environmental Sensory Based MIMO Radar Signal Process

Multiple-Input Multiple-Output (MIMO) radar has attracted attention fromscholars. MIMO radar covers broad space by beaming orthogonal or independentsignals with different antennas. After the receiving end uses the matched filter toseparate the signal from each transmitter unit, it is equivalent to the synthesis of avirtual element, thus will expand the array aperture. The precision of parameterestimation and the space covered by MIMO radar are much larger than traditionalradar system and it also expands the max number of sources. Therefore, MIMO radarhas very high practical value.To overcome the interference problem faced by MIMO radar, this paperdiscussed MIMO radar based on Electromagnetic Environmental Sensory (EES) andanalyzed the working principles of EES-MIMO.This research can be divided into fivesections: the principle and structure of the EES-MIMO radar, the electromagneticenvironment sensing method of EES-MIMO radar, the transmitted waveform design,MUSIC peaks gradient search method based on single-base EES-MIMO radar andMUSIC peaks genetic algorithm and particle swarm search method for single-base Larray EES-MIMO radar.The first section analyzed the functions of single base transceiver MIMO radar,its structure and the virtual array elements and characteristics. In order to improve theanti-interference ability of MIMO radar, this article introduces the EES-MIMO radar,and gave detailed descriptions of its structure and functionalities, thus provided atheoretical support for following researches.Complex array structures in commonly used noise spectrum estimationalgorithm are difficult to apply to MIMO radar’s electromagnetic environment andwill require an appropriate noise spectrum algorithm. The second section first analyzed the two radar interferences: active jamming and passive clutter. To solvenarrowband rifle-type interference, author used SVD method from AR model togenerate the EES-MIMO radar standard noise power spectrum estimation andanalyzed the order determination of cross-spectrum AR model. For point-frequencyinterference, author introduced cross spectrum decomposition method under thecolored noise background. This method improves estimation accuracy by expandingthe cross-correlation dimension and it is not only effective in suppressing theinfluence of the observation noise, but also reduces the amount of computation. Lastly,author proposed amplitude spectrum method to solve the interferences and thesimulation confirm the effectiveness.The third section discusses the design of EES-MIMO radar transmitterwaveform. Author first discussed the concept of orthogonal waveform design,optimizing the objective function and the basic method of waveform design usinggenetic algorithms. These discussions focus on the orthogonal Discrete FrequenciesCoding Waveform (DFCW), Orthogonal Frequency Division Multiplexing (OFDM)and Multi-Carrier frequency Phase Coding (MCPC). Lastly, author used geneticalgorithms to generate three coding waveform designs for certain range spectrum.The fourth section focused on the single-base line array of EES-MIMO radar andMUSIC spectrum peak search method. To avoid the intensive computation from DOAestimation, this article proposed a method based on one-dimensional iteration. Authoranalyzed objective function of the iteration and used first and second order derivativeto propose a three phases search method and developed a three-stage search strategy.The fifth section focused on the search method for MUSIC peak spectrum onsingle-base L array EES-MIMO radar. This article suggested two methods: geneticalgorithms and particle swarm optimization based on two-dimensional optimizationfor MUSIC spectral function. Compared to traditional MUSIC peak spectrum searchmethod, these methods require less computation and are also less complex.