Frequency Division Orthogonal Waveform Design And Processing Research For EPC-MIMO Radar

Multiple-input multiple-output(MIMO)radar emits multi-orthogonal waveforms to achieve wide or full coverage,and forms the transmitting beam at the receiving end equivalently,thereby receiving the parallel multi-beam to illuminate the completely covered space and detect targets for a continuous time.MIMO radar has the characteristics of multiple degrees of freedom,high system resolution,good anti-interception performance and strong target tracking ability,which has important theoretical and practical value.Waveform design has always been an indispensable topic in the development of MIMO radar,and has drawn scholars' extensive attention.The element pulse coding(EPC)is a new technology proposed by the research group in recent years,which further increases the flexibility of MIMO radar waveform design through joint design between element and pulse.EPC-MIMO radar still relies on orthogonal waveform design.Therefore based on the EPC-MIMO radar model,this paper focuses on the frequency division orthogonality in allusion to the problems of autocorrelation side-lobe,cross-correlation and ambiguity function in orthogonal waveform design.It uses frequency modulation agility and nonlinear frequency modulation signal to reduce the peak side-lobe level(PSL),and further with the help of intelligent bat algorithm optimize the design to improve waveforms' performance.Moreover,the alternating direction multiplier method is adopted to establish a multi-constrained optimization program of transmission-reception joint pattern,and form a beam pattern with low side-lobe finally.The main content and contributions of this article are as follows.(1)Aiming at the PSL problem in the process of multi-waveform separation in EPC-MIMO radar receiver,the processing method under different relations between frequency offset and signal sub-bandwidth is studied through analyzing the correlation function and ambiguity function characteristics of frequency division signal corresponding to different relations.When ?f is not less than Be,I design an FIR band-pass filter bank with the joint window processing to weaken the effect of the related signal components of adjacent channels.When it is less than Be,divide the array into groups and combine with different positive and negative frequency modulation rates in pairs to change the frequency distribution of adjacent channel signals,which reduces the PSL of matched filter processing and improves the performance of multi-waveform separation effectively.(2)Focus on the high cross-correlation side lobes when the ?f is less than Be,I propose a frequency division waveform method based on nonlinear frequency modulation(NLFM)agility and build an ambiguity function optimization model with the use of bat algorithm to improve the waveform performance.Compared with the LFM agility design,the crosscorrelation peak level is reduced by about 7d B through the joint design of NLFM signal and modulation agility mode,which effectively improves the performance of multi-waveform separation process.Further,in order to optimize the waveform orthogonality and improve the bandwidth resource utilization,divide the frequency offset between elements into intragroup and inter-group offset correspondingly and establish a cost function by using PSL of ambiguity function.Then according to the superiority of random search process,bat algorithm is used to randomly iterate the feasible frequency offset pairs in the solution space,which choses the optimal solution to improve the waveform correlation and enhances the beamforming performance.In addition,introduce an inertial correction factor to ensure that the algorithm converges to the global optimal result,which increases the stability of the algorithm.(3)As for the problem of MIMO radar joint transmitting-receiving two-dimensional pattern design and low side-lobe control,I study a convex pattern optimization method with sidelobe level constraints.Construct the side-lobe level constraints and the pattern weight vector as independent variables and establish the mathematical model of pattern design with some side-lobe level constraints by adopting the discrete sampling multi-constraints design.Referring to the idea of alternating direction multiplier method,optimize the iteration of lobe level and weight vector and solve the iterative variables faster.Finally,form the transmitreceive joint pattern with low side-lobe design of MIMO radar.The algorithm has the advantages of fast convergence speed and strong robustness.