Efficient Multi-Hop Communications for Software-Defined Wireless Networks
MetadataShow full item record
Software-Defined Networking (SDN) recently emerged to overcome the difficulty of network control by decoupling the control plane from the data plane. In terms of the wireless medium and mobile devices, although new challenges are introduced into SDN research, SDN promises to address many inherited problems in wireless communication networks. However, centralised SDN control brings concerns of scalability, reliability, and robustness especially for wireless networks. Considering these concerns, the use of physically distributed SDN controllers has been recognized as an effective solution. Nevertheless, it remains a challenge in regard to how the physically distributed controllers effectively communicate to form a logically centralised network control plane. Dissemination is a type of one-to-many communication service which plays an important role in control information exchange. This research focuses on the strategic packet forwarding for more efficient multi-hop communications in software-defined wireless networks. The research aim is to improve the delivery efficiency by exploiting the delay budget and node mobility. To achieve this objective, existing multi-hop forwarding methods and dissemination schemes in wireless networks are investigated and analysed. In the literature, information from the navigation system of mobile nodes has been utilised to identify candidate relay nodes. However, further studies are required to utilise partially predictable mobility based on more generalised navigational information such as the movement direction. In this research, the feasible exploitation of directional movement in path-unconstrained mobility is investigated for efficient multi-hop communications. Simulation results show that the proposed scheme outperforms the state-of-the-art because directional correlation of node movement is considered to dynamically exploit the delay budget for better selection of the relay node(s).
- Theses