Owing to the decametric wavelength, large aperture antenna array is needed for high frequency (HF) ground radar to obtain high angular resolution with conventional beam forming (CBF); for the case of compact antenna or small aperture array, spatial super-resolution algorithms are used to get satisfactory angular resolution. Wuhan University's ocean state measuring and analyzing radar (OSMAR2000) is expected to real time extract ocean surface current, wave and surface wind at medium- and long-range. To obtain long-range (i.e. 200km) current mapping, radar frequency must be selected in the lower region of the HF band, e.g. 7.5MHz. Digital beam forming (DBF) is applied with the 120m-long phased array to determine the bearings of sea echo with 151 resolution for wave and wind extraction. The angular resolution is, however, too coarse for current mapping, especially when the ocean current detection range is as far as 200km. Consequently, a super-resolution ocean surface current algorithm based on multiple signal classification (MUSIC) is developed for OSMAR2000. The relative theory basis, processing procedure and preprocessing and postprocessing associated with the algorithm are given in this paper. The comparison of OSMAR2000 measurements with the measurements from current meter and the Seasonde system shows the ocean surface current algorithm based on MUSIC for OSMAR2000 can meet the requirements of the project contract successfully.