Integrated Technology Of Forward-Looking Imaging And Detection For Airborne Radar

Airborne radar forward-looking imaging and target detection have very important applications in many military and civilian fields,such as aircraft avoidance in complex terrain,autonomous landing,precise guidance,and airdrop supply.However,conventional Doppler beam sharpening and synthetic aperture radar have a blind spot in the forward-looking imaging area.The real-aperture scanning radar is usually selected as the airborne radar to acquire the real-beam image of the forward-looking area,but it is limited by the antenna size,thus the azimuth resolution of the real-beam image obtained is low,which cannot effectively distinguish multiple targets of interest.Hence,it brings a huge challenge to target detection.Therefore,related research on airborne radar forwardlooking imaging detection integration technology is of great significance to improve the performance of airborne radar.Aiming at the problems of forward-looking imaging of airborne radar and detection of objects of interest,the main research is the integrated technology of super-resolution imaging detection based on a single real beam image.First,the echo characteristics of the real-beam forward-looking scanning radar are analyzed,and then an echo model is established.Finally,a target detection method based on low-rank sparse decomposition and an integrated method of super-resolution imaging detection are proposed.The main contents are as follows:1.Forward-looking imaging model of airborne radar is established.According to the echo acquisition process of the airborne forward-looking scanning radar,an echo signal model is derived,and then the characteristics of the echo are analyzed.At the same time,the echo azimuth convolution model is established.Subsequently,the morbidity of the convolution model is analyzed.These provide theoretical support for subsequent superresolution imaging and target detection.2.An airborne radar target detection scheme based on low rank sparse decomposition is proposed.In this scheme,the real beam image is first super-resolved to improve the azimuth resolution,so that multiple adjacent objects in the same distance unit are effectively resolved.Then the super-resolution image obtained is pre-processed using a sliding window operation and mapped super-resolution patch-image.Finally,low-rank sparse decomposition and sliding window post-processing operations are used to achieve detection of multiple targets of interest in the forward-looking area.This scheme not only effectively solves the problem of high false alarm probability of traditional detection methods,but also effectively solves the problem of multiple adjacent targets being detected as one target.3.An integrated method for forward-looking imaging detection of airborne radar based on low-rank sparse constraints is proposed.This method first analyzes the characteristics of the target of interest and background in the forward-looking scene.Then under the framework of regularization theory,the sparse target and low-rank background are used as regularization constraints to construct the objective function.Finally,the objective function is solved,while is achieving simultaneously super-resolution imaging and target detection.This method effectively reduces the problems of error transmission and improves processing efficiency.The methods proposed in this paper have been verified by MATLAB simulation and real data.The results show that the above methods can realize effective detection of multiple targets in the forward-looking area of airborne radar,and the integrated method can also achieve super-resolution imaging in the forward-looking area.