Study On Pulse Compression And Sidelobe Suppression Technology For Radar Signal Processing

Pulse compression technology as an important means of modern radar digital signal processing,the application of the technology allows the radar can improve its long-range detection performance and high range resolution performance.However the pulse compression technology based on the matching filter theory often brings higher range sidelobe interference,especially in the case of multiple targets,its severe distance sidelobe interference will cause false targets or loss of the target,thus reducing radar detection ability of the target.Therefore,how to effectively suppress the influence of the sidelobe caused by the pulse compression plays an important role in improving radar multi-target resolution and weak target detection performance.The pulse compression sidelobe suppression technology and the design of a pulse compression hardware circuit based on the segmented convolution overlapping reservation method are studied in this thesis.Based on the principle of pulse compression theory,the time domain and frequency domain implementation of pulse compression and the two methods of pulse compression using segmented convolution are discussed:overlap retention method and overlap addition method.In this thesis,the sidelobe suppression techniques are mainly discussed based on LFM signals:window function weighting method,least-squares method,and improved least-squares method.The effect of Doppler frequency shift on the sidelobe suppression performance of the window function weighted and improved least-squares method was analyzed through simulation.The improved least-squares method is proposed to solve the problem that the sidelobe suppression performance decreases when the target has a Doppler shift.The weights of the filter weights obtained under different Doppler shifts are weighted and recombined to obtain new filter weights.The new filter weight is used to perform sidelobe suppression.Compared with the unmodified method,the sidelobe suppression performance is improved when there is a Doppler shift.When the number of weighted filters corresponding to different Doppler shifts is 11,the Peak sidelobe Level performance of the improved method at higher Doppler shifts is 0.6 dB better than that of the unmodified method,and the Integrated Sidelobe Level performance is 1dB better than that of the unmodified method.And the Doppler range of sidelobe suppression can be changed flexibly by weighting coefficient.Aiming at the problem of large point frequency-domain pulse compression resource consumption and long processing cycle,the pulse-compression circuit structure in the frequency domain is designed based on the piecewise convolution overlapping reservation method.Through multiplexing one FFT module by time division,the structure completes the entire frequency domain pulse compression process in a segmented pipeline and a global serial manner,thereby reducing the hardware resource consumption.According to the length of different echo sequences,the corresponding control signals are changed and different numbers of segmented convolutions are processed,thereby increasing the processing speed.In this thesis,using Matlab to build verification platform and generate corresponding test incentives,and send it to Modelsim for simulation,after comparing Matlab model and Modelsim simulation data,it can correctly perform segmented convolution pulse compression and sidelobe suppression.The magnitude of relative error of the results is in the order of10^-3,and the Peak sidelobe Level value under multi-target is-42dB.In this thesis,the pulse compression design based on the piecewise convolution method reduces the processing delay of the same data point by at least 2048 cycles compared with the pulse compression design of not an integer power of 2 data point and pipelined architectures.