Research On Track-Before-Detect Algorithm Based On Dynamic Programming For Weak Target With Radar

The increasing diversity of the target and the escalating complexity of the environment pose significant challenges for modern radar systems in detecting and tracking the weak target characterized by low signal-to-noise ratio(SNR)or low signal-to-clutter ratio(SCR).Compared with the conventional detect-before-track(DBT)algorithm,the trackbefore-detect(TBD)algorithm is a signal processing technique better suited for detecting the weak target.Instead of employing threshold processing on single frame echo data,multi-frame echo data is directly integrated and jointly processed by the TBD algorithm,significantly improving radar detection and tracking performance for the weak target.Among various TBD implementation methods,the TBD algorithm based on dynamic programming(DP-TBD)has become the mainstream technology and research focus in the domain of weak target detection with radar,owing to its exceptional performance and straightforward implementation.Although the DP-TBD algorithm has been applied and extensively researched in the radar field at present,there are still some new problems and opportunities for performance improvement that need to be addressed,considering the diversity of radar systems,application backgrounds,and target motion characteristics.In this regard,this dissertation focuses on designing and improving the DP-TBD algorithms to enhance the detection and tracking performance for the weak target,by incorporating radar system characteristics,application background features,and target motion characteristics.The main work and contributions of this dissertation are summarized as follows:1.To deal with the problem that the conventional DP-TBD algorithm is difficult to properly select the state transition number in practical application,an improved DP-TBD algorithm based on state transition range optimization design is proposed.Compared with the conventional algorithm,the proposed one can effectively improve the detection and tracking performance for the weak target.2.For the sea detection scenario with the airborne radar,the DP-TBD algorithm under non-Gaussian distributed clutter background is researched.The DP-TBD algorithm based on K distributed clutter and the DP-TBD algorithm based on the compound Gaussian clutter with inverse Gaussian texture are proposed.Compared with the test statistics derived based on amplitude or Gaussian distributed clutter in the conventional DP-TBD algorithms,the proposed algorithms can further increase the dissimilarity between the target and non-Gaussian distributed clutter,so as to effectively improve the detection and tracking performance for weak target in the background of non-Gaussian distributed clutter.3.In the context of long-term observation,under the background of simultaneous detection of airborne and sea surface targets with the over-the-horizon(OTH)radar,a DPTBD algorithm based on kinematic constraint is proposed for the detection and tracking of the high-speed moving target by using the target motion constraints.On this basis,a DPTBD algorithm based on multi-channel is proposed,which further improves the detection and tracking performance with OTH radar for the moving weak target.4.In the context of long-term observation,a DP-TBD algorithm based on compound integration is proposed for the maneuvering target.In this algorithm,a variety of motion constraints of the target,combined with coherent and noncoherent integration methods are used to effectively accumulate the energy of the maneuvering target,so as to achieve effective detection and tracking for the maneuvering weak target.Compared with the conventional algorithms,the proposed algorithm has higher adaptability to the maneuvering characteristics of the weak target,better detection and tracking performance and more acceptable computational complexity.