The close relationship between resource deployment and specific tasks in traditional Wireless Sensor Network(WSN) leads to low resource utilization. Virtualization technology is introduced into WSN to solve the problem, and resource allocation strategy based on Semi-Markov Decision Process(SMDP) is proposed. Then, defining the state, action, and transition probability of the virtualized WSN, the expected reward is given by considering the energy and time to complete the Virtual Sensor Network Request(VSNR), and the value iteration method is utilized to maximize the long-term reward of the network resource provider. The numerical results show that the proposed resource allocation strategy can effectively improve the revenue of the network resource providers.