Research On Underdetermined Blind Source Separation Algorithm For MIMO Radar

MIMO radar antenna greatly improves the ability of radar resist destruction and target parameters estimation through a spatial distribution of the emitted by orthogonal waveforms,and matched filter reception echo.The characteristics of the existing MIMO parameters of radar reconnaissance recognition algorithm just obtain the MIMO radar,not the intra-pulse fine features,also cannot effectively improve the ability of MIMO radar jamming and anti-jamming.Therefore,this paper proposed that the underdetermined blind source separation algorithm is applied to MIMO radar signal reconnaissance recognition.According to the MIMO radar signal time-frequency characteristics,using time-frequency independent complex angle detection algorithm and adaptive hierarchical clustering algorithm to estimate the MIMO radar signal underdetermined mixing matrix,and then estimate algorithm to reconstruct the MIMO radar signals by using the norm minimization algorithm and frequency coding sequence.MIMO radar signal underdetermined blind source separation algorithm not only can be used for transmitting signal reconnaissance enemy MIMO radar to interference,also be used for processing on MIMO radar echo,it can achieve the purpose of anti-interference through separating the emission signal which does not belong to one of the MIMO radar.The research for MIMO radar signal underdetermined blind source separation algorithm is necessary and meaningful.In this paper,this algorithm is deeply studied.The main research results are as follow:?1?Do the thorough research in view of the 3 common orthogonal MIMO radar signal model of discrete frequency coding MIMO radar signal,single frequency MIMO radar signal,linear frequency modulation MIMO radar,and the main analysis include time domain,time-frequency domain signal characteristics.These kinds of signal have shown sparse characteristics,in each time slot of different emission signal occupy different frequency in time and frequency domain.The single frequency MIMO radar performance in the time-frequency characteristics of pulse period signal into a horizontal line,and the linear frequency modulation MIMO radar signal for slope with a slash,but the orthogonal discrete frequency coding MIMO radar signal performance different frequency in different time gap,and the pulse cycle frequency changes according to the coding sequence.Through the computer simulation,compared to the orthogonal discrete frequency coding MIMO radarsignal,single frequency MIMO radar signal and linear frequency modulation MIMO radar signal model is simple,therefore this paper mainly based on the orthogonal discrete frequency coding signal model of MIMO radar,which lays a theoretical foundation for the behind the underdetermined blind source separation reconstruction.?2?Proposed frequency independent complex angle detection algorithm and adaptive hierarchical clustering algorithm for orthogonal discrete frequency coding MIMO radar signal underdetermined mixing matrix estimation.Meanwhile,in the computer simulation experiment,it not only verified the algorithm to estimate the performance of mixed matrix in underdetermined condition,also make the relevant simulation in definite conditions and over-determined condition,the results show that the algorithm has good performance in these two conditions,and downward compatibility characteristics in underdetermined condition.?3?The l₁ norm minimization algorithm based on sparse model applied to reconstruct orthogonal discrete frequency coding MIMO radar signal in frequency domain,and proposed the frequency coding sequence estimation algorithm based on time domain and frequency domain signals,and also estimated frequency coding sequence of each transmitter signal,then reconstruct orthogonal discrete frequency coding MIMO radar transmitter signal on the basis of time and frequency domain signal of each restricted signal.The based on frequency code sequence estimation of signal reconstruction algorithm and direct the use of l₁ norm minimization algorithm do the comparison of computer simulation in the reconstruction of signals,according to the characteristic of autocorrelation and cross-correlation of the comparison results,illustrate the superiority of the proposed algorithm.