When the target information source is disturbed or the target is detected by passive cross-directional localization， only two-dimensional angle information exists in the single source target measurement data， so the problem of multi-source measurement data association must be solved. Aiming at the problem that the accuracy of association base on residual data association methods is limited， an association method based on angle deviation and distance variance over a period of time is proposed. The data is preliminarily correlated through the deviation of line of sight angle， and confirmed the association through the distance variance over a period of time based on the initial correlation results. The method can improve the accuracy of target data association and ensure the amount of calculation will not be too large， improve the association efficiency of two-dimensional information target data.