The multipath propagation of skywave over-the-horizon radar (OTHR) in target tracking makes the tracking filter in radar coordinate inevitably produce several tracks for a single target

and the propagation modes of the tracks are not determinate.The target state in ground coordinate can be obtained by the association of tracks

propagation modes and targets

and track fusion.Other problems of OTHR are low detection probabilities

low measurement accuracy and low data-sampling rate

which usually results in track absence in some radar dwells or the case that tracks of some modes may be never initialized.An adaptive OTHR multipath track fusion algorithm is proposed

which decides the optimal association of tracks

propagation modes and targets by the information of present time and the historical information.The mode association hypothesis of each step is modified with new data cumulating.The simulation results show that the algorithm of this paper can greatly decrease the state estimation error

compared with the multihypothesis fusion of multipath OTHR tracks algorithm.