Controller Failure Recovery Strategy for Software Defined Flying Ad-hoc Network Based on Latency and Node State

In order to overcome the problem of network paralysis caused by control node failure in current software defined flying ad-hoc networks (SD-FANET), this paper proposes a backup control node selection method based on latency and node state. This method combines the characteristics of network node energy limitation and link instability, and sets the corresponding energy detection method and timeout detection mechanism to judge whether the control node is invalid. After determining the control node failure, the node availability is calculated according to the network feedback parameters such as link latency, joint strength and node energy. Based on this, the backup control node is selected, and the backup control node completes the networking again according to the backup data received from the control node, so as to complete the failure recovery. The simulation results show that, compared with the existing election strategy, the proposed method can effectively reduce the number of messages and recovery latency in the process of recovery failure, and can quickly complete the failure recovery with lower network overhead; moreover, compared with the random selection algorithm, the failure recovery rate of the algorithm is improved by more than 38%, which can ensure the reliability and practicability of the network to a greater extent.