Analysis Of Data Fusion Algorithm For Distributed Multi-ship Based Surface Wave Over-the-horizon Radar

Distributed Multi-ship based Surface Wave Over-The-Horizon Radar (DMS-SWOTHR) is the use of natural equipment resources, which are formations of thefleet contains multiple coexisting warships, construct a distributed mobile shipbased surface wave over-the-horizon radar system. This system is not only make upfor the technology limitations of single ship based surface wave OTH radar, but alsocan greatly improve the overall operational performance, such as the systemdetection ability, anti-jamming performance, target detection accuracy and thesurvival ability, etc. As an important works in the aspect of a DMS-SWR radar dataprocessing, data fusion has great effect on the overall tracking performance of radarsystem. This paper focuses on the data fusion algorithms, the application scheme oftarget tracking algorithm, track correlation algorithm and track fusion algorithm inthis topic background, and studies how much the layout of distributed effect theoverall tracking precision of the system.First of all, the paper deduced and established a highly flexible trackingcoordinates for DMS-SWR specific data processing environment. Considerd thatHFSW radar exhibits poor azimuth resolution, high non-linearity, and significantfalse alarm rate, the Joint Probablistic Data Association (JPDA) with the UnscentedKalman Filter (UKF) is proposed. As simulation shows, the overall filterperformances of the JPDA/UKF algorithm adopted in this paper are better than thatof the JPDA with the Extended Kalman Filter (EKF).Second, focusing on the problem of track correlation, this paper presents atrack association based on curve fitting algorithm, which does not appear leakageassociated situation, also not need ambiguity processing, simple in principle, easy toimplement. The simulation results show that the correct correlation rate of thismethod is much higher than that of traditional methods under certain condition.Afterwards, study on current distributed wide used fusion algorithms, such assimple convex combination fusion algorithm, the weight covariance fusionalgorithm, no feedback fusion algorithm. Simulation of the three algorithms showsthat those algorithms performance only has a slightly different for DMS-SWRsystem. Also found that the fusion algorithms can effectively improve the trackingaccuracy of the system.Lastly, research on DMS-SWR system containing2,3,4warships shows thatthe increased fusion number of warships to some extent can improve the trackingaccuracy of the system. There are some influence factors effect on the trackingprecision of DMS-SWR system, which composed of the same precision radars, inthis study: the father distance between the fleet and the target is, the worse the system tracking accuracy obtains; the better the degree of position improvementcoused by the formation of the ship moving is, the higher the system trackingprecison obtains; the tracking precision of disorder formation better than that ofparallel formation. In the study of the DMS-SWR system consisting of differentprecison radars, the results show that the use of high precison radar in DMS-SWRsystem can improve the tracking accuracy under conditions of the radars.avergeaccuracy same.