In order to the problems of poor update quality and low survival rate of cluster head nodes in the deployment process of mobile sensor networks,a cluster head update algorithm for mobile sensor networks based on matching trust mechanism is proposed.First,K-means algorithm is introduced to complete the initial clustering of the network by using the square root function of error to quickly locate the cluster center and improve the clustering speed.Then,considering the three factors of the residual energy of the candidate cluster head,the Euclidean distance bet ween the candidate cluster head and the current cluster head,and the total number of nodes within the coverage range of the candidate cluster head, a cluster head update method based on matching trust is designed.Each factor is evenly distributed according to the weight value,and then the node with the highest trust weight value is used as the candidate cluster head,so as to elect the node with higher survival quality.Simulation results show that the algorithm in this paper has higher network stable running time,higher cluster head node survival rate,and lower node failure probability.Among them,the stable running time of the network has been improved by over 80%,the survival rate of cluster head nodes has been maintained at over 90%,and the probability of node failure is also low,which has obvious advantages.