Research On Waveform Optimization Design Of MIMO Radar

Since collocated MIMO radar has the advantage of waveform diversity,it can increase the degree of freedom of transmitting signal,switch the working mode flexibly,and enhance the detection accuracy of radar.Therefore,it is widely concerned by scholars.For synthetic aperture radar(SAR),the spatial diversity can be obtained by orthogonal waveform design,which can improve the detection performance and positioning accuracy of the target.In addition,with the development of millimeter wave radar and 5G technology,scholars found that radar and communication system have many similarities in technology.Some specific application scenarios,such as automatic driving,need radar detection and information communication separately.People can fuse radar detection signal and communication signal,and radar communication integrated waveform design is a research hotspot.Therefore,this paper mainly focuses on the waveform optimization design of MIMO radar and the integrated waveform design of radar communication.The main innovations are shown as below:1)The traditional MIMO OFDM waveforms based on chirp signal,which only contains two orthogonal signals with small frequency offset at the initial frequency,and the signal processing flow is complex,without considering the system interference in practical application conditions.A new design scheme of four-way MIMO OFDM waveforms based on chirp signals is proposed.First of all,the original LFM signals are compensated by phase compensation,FFT transform,and zero interpolation,which make each signal completely orthogonal in frequency domain,and the initial frequencies are the same,which is conducive to the generation of signal in practical application.At the same time,the time-domain echo signals after spatial filtering are processed by means of zero interpolation and down-sampling,which simplifies the demodulation process and extracts the target information of echo signal.In order to avoid the influence of thermal noise and spectrum shift on the autocorrelation and cross-correlation sidelobes,small time intervals are added between the periodic LFM sub-pulses in a pulse repetition period.Through the experimental simulation analysis,the MIMO OFDM chirp signals proposed in this thesis have lower autocorrelation and cross-correlation sidelobes,while expanding the number of orthogonal waveforms,and maintaining the same radar detection performance as the traditional MIMO OFDM chirp signal.2)The traditional MIMO OFDM chirp radar signals based on multi-carrier frequency have high autocorrelation and cross correlation sidelobes,which can produce false targets or cover up some small targets,MIMO OFDM radar signals based on multi-carrier PNLFM sub-pulses are proposed.First of all,the traditional orthogonal signals only contains LFM sub-pulses of up and down frequency modulation,and the modulation frequencies are the same or similar,resulting in a large number of high interference sidelobes produced by the cross-correlation function of the sub-pulses of the orthogonal signals.Therefore,this thesis proposes a PNLFM waveform,which is composed of three segments.The first segment and the third segment are LFM sub-pulse,the second segment is NLFM sub-pulse.Each PNLFM waveform can change the time width and bandwidth of each sub-pulse by their controllable variables,which increases the freedom of MIMO radar signals.At the same time,the time widths of the sub-pulses are changed while the bandwidths of the sub-pulses are kept constant.Then,the controllable variables of each sub-pulse are optimized by GA algorithm.The optimal MIMO OFDM radar signals of multi-carrier PNLFM sub-pulses are obtained,which effectively suppresses the interference sidelobes.Through a large number of simulation experiments,it can be found that the MIMO OFDM PNLFM radar signal proposed in this thesis has lower autocorrelation and cross-correlation sidelobes.3)In practical SAR systems,multi-carrier MIMO OFDM radar signals based on PNLFM sub-pulses can't maintain low correlation sidelobes,due to the interference of orthogonality heating noise and time delay.This thesis proposes single-carrier MIMO radar signals based on multiple PNLFM sub-pulses,which greatly increase the number of orthogonal waveforms.The sub-pulses of this signal are composed of different types of PNLFM signals.Each sub-pulse has its own control variable,which greatly increases the number of orthogonal waveforms and the freedom of the MIMO radar signal.The GA algorithm is used to optimize each control variable to obtain an optimized waveform,which effectively reduces the autocorrelation and cross-correlation sidelobes of the MIMO radar signal,and has good orthogonality.Then,the structure of the PNLFM sub-pulse is further analyzed,and a MIMO radar signal based on a single carrier frequency discontinuous PNLFM sub-pulse is proposed.The PNLFM sub-pulses of the orthogonal waveform of this signal have a discontinuous time-frequency diagram structure,which minimizes the interference sidelobes caused by the sub-pulse cross-correlation function.Because the waveform is composed of PNLFM sub-pulses of the same kind,the number is relatively small,and it is suitable for MIMO radar systems with fewer array antennas.The swarm particle(PSO)algorithm is used to optimize each control variable to speed up the convergence of the model and obtain the optimized waveform.Simulation experiment results show that with the increase of the time-bandwidth product,the model effectively suppresses the correlation side-lobe interference,and obtains MIMO radar signals with lower auto-correlation and cross-correlation side lobes.4)In the conventional integration system of radar and communication with automobile,there is only one restriction condition about the total transmitting power,which makes the integration signals distorted easily.In addition,the conventional integration waveform of radar and communication which takes the trade-off between multi-user interference and radar reference signal similarity as the objective function.Thus,the novel integration system of radar and communication take the same objective function,with the constraint condition of total transmitting power and PAPR proposed.The constant envelope constraint is a special case of PAPR constraint.In the integration waveform design of radar and communication,the optimization model is quadratic constrained quadratic programming(QCQP).The optimization model can be obtained by using the semi-definite relaxation(SDR)technology.The integration signals of radar and communication generated by our optimization model are compared with the integration waveforms of radar and communication generated by the total transmit power constraint.They have almost the same radar detection and communication performance.Our novel waveform PAPR value is small,in the real transmitting process,they are transmitted through the nonlinear amplifier without serious distortion,and the radar detection performance and power utilization ratio of transmitters can be improved.At the same time,the radar communication integrated signals generated under PAPR constraints have better radar detection and communication performance than the radar communication integrated signal generated under constant envelope constraint.