Study On Incorporated Targets Detection And Tracking Method For High Frequency Surface Wave Radar

High Frequency Surface Wave Radar (HFSWR) has the characteristic of long range detection, real-time as well as working constantly for all the time. It plays an important part in the vessel target detection within the 200 nautical mile exclusive economic zone of our country. Traditional HFSWR targets detection method is track-after-detect (TAD). This method first gets the targets’points through Constant False-Alarm Rate (CFAR) detection of radar echo signal, and then carries on targets tracking. But CFAR detection often causes false alarm or missing detection of dim targets. Moreover, the targets tracking based on detected points usually faces the difficulty of points association of adjacent time. However, track-before-detect (TBD) is a new targets detection method. Unlike TAD, TBD does no detection on single frame of echo data. On the contrary, it accumulates multi-frames of echo data to fully retain the effective information of the original data, and then carries on targets tracking by utilizing the relevance between adjacent frames. When tracking is done, the targets’points are also detected almost simultaneously, thus realizing the incorporated target detection and tracking.This thesis focuses on the study of dynamic programming based track-before-detect (DP-TBD) method for HFSWR targets detection and tracking as well as the verifying of the method. The main work includes:1. Proposes an improved DP-TBD method for HFSWR targets detection. This thesis carries on a thorough research and analysis on the common DP-TBD method. Then the characteristics and drawbacks of the method are concluded. The drawbacks involves large amount of calculation, difficulty in threshold setting and converge of tracks. On this basis, this thesis proposed an improved method. The improved DP-TBD method adopts CFAR pre-processing step to reduce the amount of calculation and avoid the difficult of threshold setting. Then the joint toward and backward track searching and association method is proposed to solve the problem of track converge. The verifying results based on measured HFSWR data show that, the proposed improved DP-TBD method can solve the problems of common DP-TBD method effectively. At the same time, the tracking continuity is obviously better than traditional TAD method.2. Carries on a research about the HFSWR tracking results verification methods based on AIS data. When using AIS to verify the tracking results of HFSWR, the time synchronization interpolation of AIS tracks is needed. If there is a long time interval between two adjacent AIS messages sent by one vessel, the traditional linear interpolation for AIS may cause large deviation. To solve this problem, this article proposes a new AIS track interpolation method. This method forecasts the vessel’s positions during the interpolation time period by taking into account the dynamic information of AIS, such as speed, course and so on, and then forms the interpolation track of the moving vessel. At last, the proposed method is verified by measured AIS data. The results show that, comparing with linear interpolation, the track interpolation result of the method proposed by this article is closer to the real vessel track during the interpolation time period.3. The tracking performance of the improved DP-TBD method is verified via a cooperative vessel target detection experiment. The comparison and analysis results show that the cooperative target’s track produced by the proposed improved DP-TBD method is much close to its AIS track. Besides, there is only a small radical velocity and range error between the HFSWR track and AIS track.