This article considers a centralised data fusion system, in which the measurements of the local sensors are time-stamped, and then transmitted through the network to the fusion centre. The system will suffer measurements delay or loss due to the unreliability of the network. Based on the Kalman filter and information filter in a single channel, a centralised data fusion algorithm with buffers is proposed to solve this problem. A probabilistic metric to evaluate the performance of the system is presented. Simulation results show the effectiveness of the proposed method.