The essential proteins in protein networks play an important role in complex cellular functions and their evolution. Therefore, searching essential proteins in protein networks can help to explain the structure, function and dynamics of basic cellular networks. The existing dynamic protein network regards the protein situation at all times as the same, but in fact, the role of proteins varies at different times. In order to improve the accuracy of essential protein searching, an improved H-index algorithm based on attenuation coefficient method is proposed in this paper, which incorporates the neglected node information into consideration to improve the accuracy of essential protein searching. The experiments show that the essential proteins found on the basis of this model are more effective than other similar methods.