Research On Knowledge-based High Frequency Surface Wave Radar Target Tracking Methods

High frequency surface wave radar(HFSWR)has attracted great interests for target detection and tracking by virtue of their low-cost,over-the-horizon coverage capability and continuous-time mode of operation.It can obtain continuous trajectory of vessel targets by target tracking process and becomes a dominant tool for maritime surveillance.Unfortunately,HFSWR also exhibits many shortcomings that need to be taken into account when tracking targets independently.For example,low detection precision and sea clutter disturbance will result in inaccurate targeting and even track loss.Therefore,it is difficult to track targets accurately only relying on the radar's own detection data.In order to overcome these limitations,this thesis studied the knowledge-based target tracking methodology of HFSWR and mainly finished the following innovative work.Firstly,in consideration of the disturbance of the first-order sea clutter when tracking vessel targets,some targets cannot be detected under single frequency.These missed contacts will result in target missing and track fragmentation.After analyzing the relation between target tracking results and radar operating frequency,as well as the traditional target tracking methods of HFSWR based on dual frequency auxiliary information,a new target tracking method based on time-devison dual frequency information is proposed.Target tracking is performed by setting different tracking frequencies at different times.Simulation and real radar tracking results show that the proposed method effectively solved the problem of target missing caused by blind zone.Secondly,in view of low range and azimuth resolution of HFSWR,the tracking results will be inaccurate and the measured track will deviate from the target's true trajectory.A track correction method exploiting Automatic Identification System(AIS)information is proposed to improve the tracking performance.Then range and azimuth correction models are established by statistical regression methods during HFSWR and AIS synchronous tracking period and the measured range and azimuth are corrected via the established models to improve the tracking accuracy during independent tracking period of HFSWR.The results of track correction experiments show that the proposed method can greatly reduce the deviation of the track and improve the track quality.Finally,in order to accurately estimate process noise variance and measurement noise variance in the state estimation process,a noise variance estimation method based on AIS information and innovation is proposed.By this method,the noise variance estimation scheme is designed for two situations where the targets may appear.Experimental results by simulation and real target data demonstrate that the proposed method can obtain accurate estimation for both measurement and process noise variances.Moreover,with the estimated parameters,higher quality track can be obtained.