Radar Signal Processing Key Technologies In Low-slow Small Target Detection

With the rapid development of stealth aircraft and unmanned aerial vehicle technology,the detection of low-slow small target has put forward higher performance requirements for radar in recent years.From the viewpoint of radar signal processing,the detection of low-slow small targets is studied systematically in the thesis,which further improves the detection performance.However,due to their weak echo,low-slow small targets are very likely to be submerged in clutter,which means that extracting the target component from the clutter effectively is really crucial.Firstly,the transmitting and echo signals of radar are introduced.The characteristics of echo and clutter for low-slow small targets are analyzed,while the clutter spectrum characteristics in the static and moving platform are also given.Then,the filter design methods are introduced for clutter suppression.The clutter suppression performance is verified by numerical simulations.On this basis,an effective filter design method is proposed.This method adopts the optimized FIR filter design method at low freque ncy,while the MTI cascaded FFT filter design method is used at the intermediate frequency,which shows effective clutter suppression performance at low frequency and has small computation complexity.Next,we study more advanced algorithms such as empirical mode decomposition(EMD)and variational mode decomposition(VMD)in radar signal processing.The theoretical analysis and measured data processing results show two algorithms have excellent clutter suppression performance.Finally,the thesis gives angle estimation methods in low-slow small target detection for mechanical scanning radar.It introduces two conventional angle estimation methods.Considering the angle estimation precision under strong clutter background and the coupling of slow time and angle,a new algorithm is proposed to further improve the precision based on MTD pulse accumulation and sliding window processing.The superiority of the new algorithm is verified by computer simulations.