Each group member should upload a copy of the proposal.
The research question that we attempt to solve in this study lies in the efficiency of routing protocols in a drone swarm environment. We use a drone swarm because of the dynamic nature of the network that it provides, as opposed to other DTN scenarios that may or may not provide the same level of robustness to networking challenges to include short connectivity times, overhead management issues, and power management issues to name a few. In particular, we aim to answer the following question: “What DTN routing protocol most efficiently enables drones to communicate when operating in a swarm environment?” by testing Epidemic, Vector, Centroid, GAPR, and GAPR2. We will attempt to answer this question by executing multiple simulations, compiling the data, developing a way to objectively compare the results, and presenting this information in a clear and concise manner.
Greasy is a NPS-developed tarpit currently capable of operating via IPv4 or IPv6 Internet. There is an associated IPv4 software defined network (SDN) controller module developed to work in conjunction with Greasy, but IPv6 functionality does not currently exist. The project objective is to design a working IPv6 SDN virtual environment using Mininet and use it to emulate the performance of Greasy in that environment.The project’s scope is specifically limited to IPv6, and the size of the network will be academically set at one link, two links, then some (small) arbitrary number of links. The primary gain for academia will be affirming Greasy’s tarpit capability by comparing the operation of Greasy on IPv6 with and without the use of the SDN functionality.
This project examines whether segmenting a large MANET into smaller networks and connecting ‘edge’ nodes into an overlay MANET would increase throughput, decrease control messages, and increase effective range on wireless networks.
In an afloat environment, the greatest bottleneck with regards to network communication is the satellite link off the ship to a ground station. This link measures around 2Mbps of bandwidth per ship, and is utilized by 270-350 users. Furthermore, the employed system is designed with priority traffic in mind, reducing the available bandwidth to users with lower priority as needed to support mission requirements. Our objective is to optimize the goodput of this link by minimizing the amount of overhead that is sent.
In rural areas, establishing vehicular networks can be difficult because of limited communication infrastructure and the small number of vehicles on the road. This project aims to simulate such an environment in order to determine the optimal distance between roadside units (RSU) so that vehicles can successfully receive network services such as entertainment and traffic conditions.
Several SDN controllers currently exist to enable the 'new' network SDN paradigm. While work has been done to compare SDN controllers and rank based on controller features properties (e.g. rest API, productivity, documentation, modularity) and even some network performance metrics albeit in Cbench, our goal is to determine controller selection based on network performance parameters (e.g. throughput, latency) but using a different emulation environment than Cbench. We will use four well documented controllers, OpenDaylight, ONOS, Ryu, and Floodlight, to evaluate which controller is the best performer for a SDN implementation.
Project summaries are Copyright © by their respective authors.
| Day | Time | Group |
|---|---|---|
| Wednesday, 6/13 | 08:00–08:15 | Blevins, Felder, Wu |
| 08:18–08:33 | Blauwkamp, Hunter | |
| 08:36–08:51 | Kim, Mooring, Weitzel | |
| 08:54–09:09 | Brown, Davis, Hopchak | |
| 09:12–09:27 | Reynoso, Salazar, Shaw | |
| 09:30–09:48 | Amos, Tollefson, Walton |