Relative position is very important tactical information in large-scale joint military operations, especially in some cases that the members of the datalink network may not be able to obtain their position through absolute positioning, such as the environment with strong interference or environment with fewer visible GNSS satellites. The EKF (Extended Kalman Filter) algorithm based on INS/TOA/BA is proposed in the paper, members of network can process INS/TOA/BA data in two-dimensional plane timely, and height is positioned with height filter. A simulation of TDMA relative navigation network is carried out to verify the performance of proposed algorithm. The results show that compared with general WLS (Weighted Least Squares) the proposed algorithm can realize the real-time processing of TOA information in TDMA datalink network and achieve a satisfactory accuracy.