Abstract: To solve the problem of station identification, the basic theory of wavelet frame is introduced to make the smart electricity meter use the same wavelet framework, respectively to conduct pyramidal decomposition for the voltage data gathered themselves, and to frame the signal decomposition after calculating the distribution of energy. The position of sampling signal mutation is determined by differential calculation, the broadband carrier communication network of the existing electricity information acquisition system is used to make the smart electricity meters judge the results interactively. Finally, they can identify their own stations through the similarity of the mutation positions. Theoretical analysis and practical application show that the novel technique makes full use of the advantages of wavelet framework on low computational complexity in signal analysis, low memory data retention and small communication load between nodes, and the high accuracy rate of identification results. It provides a new solution for the management department which can remotely identify a large number of smart electricity meters at the same time without introducing new hardware equipment.