We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Topology management for flying ad hoc networks based on particle swarm optimization and software-defined networking.
- Authors
Pasandideh, Faezeh; Silva, Tulio Dapper e; Silva, Antonio Arlis Santos da; Pignaton de Freitas, Edison
- Abstract
Flying Ad Hoc Networks (FANETs) are composed of a set of high mobility flying nodes, such as unmanned aerial vehicles (UAVs), connected in an ad-hoc manner and collaborating to perform specific tasks or to achieve specific goals, such as providing connection to other nodes on the ground. The high mobility degree of UAVs, and the possible connected users on the ground, might cause fast and frequent changes in the network topology. Hence, the topology management adaptation to the UAVs' movements is required to reduce UAVs' mobility negative effects on the communication, and to improve the overall network performance. Observing these needs, this paper proposes a Software-defined networking (SDN) based manageable topology formation to construct a more resilient and manageable UAV formation. This novel proposal considers a set of graph theory concepts for network evaluation to guarantee user connectivity, alternative transmission paths, and lower possible amount of nodes being points of failure, as a consequence. Also, the spring virtual force method is applied by using attractive-repulsive forces among nodes to accomplish the following objectives: to impose safety distance gaps for collision avoidance; to provide sufficient communication link distance for proper link quality; and to maximize area coverage for enabling end-user mobility. Finally, the Particle Swarm Optimization (PSO) algorithm's particle selection procedure is proposed to maximize the number of interconnected nodes. Simulation results show that the proposed solution can correct routing policies and reestablish connections in every occurrence of failure. The results also indicate that the considered packet loss was significantly lower compared to the state-of-the-art, achieving results from 10% to 80% lower in the performed experiments, as a higher number of packets were delivered within the required delay limit.
- Subjects
PARTICLE swarm optimization; SOFTWARE-defined networking; ROUTING algorithms; TOPOLOGY; CHARGE carrier mobility; GRAPH theory; NETWORK performance
- Publication
Wireless Networks (10220038), 2022, Vol 28, Issue 1, p257
- ISSN
1022-0038
- Publication type
Article
- DOI
10.1007/s11276-021-02835-4