In target tracking system, the measurements from different sensors might arrive at the fusion center out of sequence because of the different communication delays, and thus result in the Out-of-Sequence Measurement (OOSM) problem.Usually there are multiple OOSMs in the actual working process of the system.Aiming at this problem, the common situation is classified, and Dl algorithm is proposed within the forward prediction filtering framework.The algorithm can update the state estimation and the covariance matrix on each moment of OOSM fusion period, and deal with multiple OOSMs with arbitrary steps of delay.In addition, the equivalent measurement information filter is deduced for single OOSM without requiring the inverse of the state transition matrix and the discrete model of process noise.Simulations verify the precision and effectiveness of the proposed algorithm.