Abstract:RSU deployment is affected by road environment characteristics, traffic flow characteristics and costs et al. The research object is the RSU deployment strategy of one-way four-lane highways. An improved KMeans vehicle clustering algorithm based on highway environment characteristics and vehicle characteristics is proposed to ensure the reliability of the network and reduce RSU deployment cost. In addition, compare improved KMeans clustering algorithm with the classic KMeans algorithm in three aspects: network residual energy, number of nodes alive and end-to-end delay. The results show that the proposed algorithm has better performance; Then put forward the RSU deployment scheme adopted in this research; Finally, study the RSU deployment: analyze the relationship between vehicle density, RSU communication radius, vehicle connectivity rate and RSU deployment interval through MATLAB. Obtain the best RSU deployment interval and communication radius. Thus providing a basis for highway RSU deployment.