The aim of this paper is to propose, develop and test multi-radar multi-target tracking algorithm for maritime surveillance at over-the-horizon (OTH) distances. The proposed algorithm utilizes a weighted pure minimum mean square error (MMSE) algorithm for multi-radar data fusion and adaptive alpha - beta tracking algorithm for target tracking. The developed data fusion algorithm calculates the weighted center of mass of all targets observed by OTH radars (OTHRs) within a gate radius of predicted position, and uses it for a track update process. During the process a weighting factor which will be used in an adaptive alpha-beta tracking algorithm is derived. The weighting factor as well as alpha and beta factors are updated in each observation period and depend on all targets, which have fallen into a gate radius during the track update process. The algorithm is designed, implemented, and tested in an environment consisting of a simulated network of two OTHRs with significantly overlapping surveillance areas.