Leader Election Algorithms (LEAs) play a vital role in distributed systems. The leader organizes and synchronizes processes in distributed systems and communication networks. The leader is prone to fail which in turn makes the network inconsistent. LEAs solve the leader failure problem by electing a new one. In this paper, a new LEA is proposed to solve the leader failure in hive networks. The proposed algorithm elects one node with the highest priority to be the new leader. The algorithm is evaluated theoretically by calculating number of messages and time steps required to elect a new leader. We show that the algorithm needs O(n) messages in O(?n ) time steps in the best case, while it needs O(n1.3) messages in O(?n) time steps in worst case.