Wireless sensor networks（ WSNs） contains a lot of redundant data

these data consumes too much energy of network; current data fusion algorithm doesn’t consider data links between each other inadequately

and fusion precision is also need to be improved. Aiming at problems

put forward a kind of energy-efficient data fusion algorithm based on improved support degree. The algorithm fuses datas in sensing nodes of network according to threshold value and introduce the concept of self-support degree combined with grey proximity theory to improve support degree functions

apply the improved support function to sink node to calculate the final data fusion estimated value. The Matlab simulation results show that the algorithm can save energy about 27. 87 %

absolute error is about 0. 98 ℃

achieves the purpose of energy saving and fusion precision improving.