Study On MIMO Radar Orthogonal Waveform Design

MIMO radar has many advantages such as high system freedom,strong anti-interception ability,good doppler resolution,and outstanding weak target detection ability.Orthogonal waveform design,as an important subject in the field of MIMO radar,has drawn extensive attention from researchers.The performance of MIMO orthogonal waveforms is usually reflected in mainlobe-peak-to-sidelobe-peak-ratio,orthogonality,and mainlobe width.As the ratio of mainlob peak to sidelob peak increases,the capability of the system for weak target detection capability would be enlarged.The orthogonality affects the signal recovery effect of matched filtering,while the mainlobe width reflects the range resolution of the radar system.This thesis focuses on the design of orthogonal phase-coded waveforms and orthogonal frequency division multiplexing nonlinear frequency modulation signals.Aiming at the problems of coarse granularity and slow convergence speed of the existing GA-SQP algorithm,the individual segment crossover operation and mutation operation were refined to optimize the algorithm's convergence efficiency;In order to simply the complicated procedures and reduce the intensive computation load of the existing NLFM signal design method,an OFDM-NLFM signal design method with an explicit frequency function is improved,which greatly simplifies the design steps while maintaining the performance.The main work of this thesis is presented as follows:In the first part,the signal transmitting&receiving model and signal processing of the MIMO radar are briefly introduced.Firstly,the advantages of MIMO radar over PA radar are summarized.Secondly,we analyzed the impact of MIMO radar signal processing order on resource consumption and system composition.Finally,the definition of ambiguity function is given with four proved main properties.The second part focuses on the design of orthogonal phase-coded waveforms.Firstly,we introduced the signal model,waveform performance indicators and common optimization criterion of orthogonal phase-coded waveforms.We briefly described the steps of common waveform design methods.The GA-SQP algorithm has both global search and local optimization capabilities.This thesis improves the crossover and mutation operations of the original GA-SQP algorithm,refines the gene-locus selection from individuals to chromosomes,and improves the convergence efficiency of the algorithm.Finally,in order to further reduce the sidelobe level,a waveform optimization method designed with mismatch filter is used.Although the singal-to-noise ratio is decreased,we got huge improvement of mainlobe-peak-to-sidelobe-peak-ratio.In addition,the effects of changes in signal number,code length,similar constraint,and mismatch filter length on mainlobe-peakto-sidelobe-peak-ratio are compared and analyzed,which provides a basis for selecting appropriate waveform design parameters.In the third part,this thesis mainly studies OFDM-NLFM signal design method.Firstly,we introduced the design method of NLFM signal based on the principle of stationary phase,and analyzed the performance indicators of the designed NLFM signal.Utilizing OFDM technology,we completed the design of OFDM-NLFM signal and verified its orthogonal performance.Secondly,signal form of NLFM based on combined window function is used to achieve a balance between the mainlobe width and mainlobe-peak-to-sidelobe-peak-ratio by adjusting parameters.Considering the inevitable error,the calculation complexity and the lack of explicit frequency function of the design method based on the principle of stationary phase,we improved an OFDM-NLFM signal design method with a simple form and adjustable parameters.Simulations show that the signal is not sensitive to doppler frequency,and has good orthogonality and high mainlobe-peak-to-sidelobe-peak-ratio.Finally,in order to further improve the feasibility of the project implementation,the designed signal is simplified by segmented LFM and discretization.The simulation results show that the sidelobes of the simplified signal are still at a low level.