Abstract:In view of the uncertainty of data collected in multi-sensor networks,a multi-sensor data fusion algorithm based on D-S evidence theory is put forward.The algorithm is divided by two steps:homogeneous data fusion and heterogeneous data fusion.Firstly,we get some eigenvalues from multiple sensors' data.By calculating the distance on homogeneous data,we get the trust functions and set threshold value in order to eliminate outliers.After eliminating outliers,we can initially fuse the data.Secondly,we can calculate the distance between heterogeneous data and the eigenvalues of each level.we calculate the distance between the function on supporting degree and getting basic probability assignment.According to the evidence theory,the fusion result will be obtained finally.The simulation results show that this method can effectively deal with the conflicts in D-S theory and obtain more accurate fusion results.