M398 applications of wireless sensor networks rely on the precise position of sensor nodes, so node localization is one of the most important issues in wireless sensor networks. Due to its simplicity, DV-Hop has drawn much attention among all the node localization algorithms. However, the DV-Hop localization algorithm uses the product of the hop count and the average hop distance to estimate the distance and uses the maximum likelihood estimation method to locate, which lead to a large error. Therefore, we use DV-Hop as a basis for positioning algorithm and improve its accuracy from two aspects. On the one hand, we introduce reference hops to correct the average hop distance between beacon nodes. On the other hand, we introduce the cycloid equation to calculate the distance from the unknown node to the beacon node. The MATLAB simulation results show that the improved algorithm has a lower error rate than the traditional DV-Hop algorithm and can be better applied to the actual scene.