Controller placement in software defined FANET

To deal with the problem of complex changes of topology in UAVs (Unmanned Aerial Vehicles) network and the increasing need of data transmission, the combination of self-organizing and SDN (Software Defined Network) is a usefull paradigm. However, the control node's position in SDN will influence the transmission of command between controller and relay nodes, thus determining the quality of communication. In this context, we present a network construction, named SDN-FANET (Flying Ad-Hoc Network). We formulate the problem as a classical network center location model. To solve this problem, we propose a particle swarm algorithm based on the graph theory. Simulation results show the algorithm performs better in terms of accuracy and stability, compared with several benchmarks. Particularly, with the initial particle position optimized, the proposed algorithm converges faster.

Automated summary

To address the complex topology changes in UAVs network and increasing data transmission needs, a useful paradigm is the combination of self-organizing and SDN. However, the position of the control node in SDN impacts the command transmission between the controller and relay nodes, determining the communication quality. In this study, a network construction called SDN-FANET is introduced, formulating the problem as a classical network center location model. A particle swarm algorithm based on graph theory is proposed to solve the problem, and simulation results demonstrate its better performance in terms of accuracy and stability compared to benchmarks, with faster convergence achieved through optimized initial particle positions.