Research On Multi-frame Tracking Before Detection Technology On Airborne Early Warning Radars

The early warning aircraft is an important symbol of the military's informatization and the nerve center of the entire combat system of modern warfare.Placing the early warning radar on an airborne platform has very important military significance such as increasing the detection area and improving the survivability of the radar.Detecting and tracking targets is the primary task of early warning radar,so the algorithm used by radar will greatly affect its effect.However,the performance of traditional detection before tracking algorithm will drop in low signal-to-noise / noise ratio scenarios.Multi-frame tracking before detection(MFTBD)algorithm,which has been proposed in recent years,can directly process consecutive frames of raw data,and use the motion characteristics of the target to accumulate along the time dimension to enlarge the differences between target and the background(Noise or clutter).The improvement of detection and tracking performance has important strategic significances for airborne early-warning radar,such as early detect and completeness of tracks.However,most of the existing research work focuses on the characteristics of the target and clutter under the airborne radar system,without considering the impact of the motion of the aircraft on the actual use of the MFBTD algorithm.This thesis focuses on the corresponding research on the airborne MFTBD algorithm.The main work is as follows:1.The algorithm model of the stationary platform radar is studied,and the principle of using the dynamic programming method to implement the MFTBD algorithm is derived.By analyzing the traditional detection and tracking algorithm of airborne radar,the correlation between measured data in different coordinate systems is given.2.For the problem that the radar model of the stationary platform cannot describe the multiple measurement information on the airborne platform,an algorithm model of the airborne early warning radar is established.In addition to the measurement information of the airborne radar,the model can also describe the information of the aircraft,which lays the theoretical foundation for the design of airborne MFTBD algorithm.3.Aiming at the problem that the airborne radar measurement planes overlap with each other at different times,and the coordinate system where the measurement data is located is not aligned,an airborne multi-frame tracking before detection algorithm based on the minimum coverage airspace is proposed.The algorithm will map measured datas at different times in the same coordinate system to meet the data form required by MFTBD algorithm.4.Aiming at the problems of redundant calculations and inaccurate measurement information caused by the failure to meet the "one-to-one mapping" relationship,an airborne multi-frame tracking before detection algorithm based on motion compensation search is proposed.The algorithm is designed under the antenna reference system,which can realize the non-destructive use of measurement information,and complete the merit function accumulation of the MFTBD algorithm with high efficiency and high accuracy.Simulation experiments verify that both algorithms can implement the MFTBD algorithm on an airborne platform,which significantly improves the radar's discovery probability for weak targets.And through comparison,it shows that the compensation search algorithm has higher detection performance and tracking accuracy.