Description
(1) Dynamic routing mechanism design with focus on throughput
Figure 1 ‑ Generic network of routers
Figure 2 ‑ Same topology as Figure 1, but now all links are wireless, and each device can be both source of traffic and also act as a relay
Create and simulate a new routing strategy that maximizes the overall throughput of a mesh network. Throughput is affected by many factors that should be considered, such as nodal processing delay, overloaded buffers, loss, etc. The more realistic assumptions you can make for your network, better it is.
(2) Dynamic routing mechanism design with focus on energy conservation
Figure 3 ‑ Wireless Sensor Network example
Create and simulate a sensor network. The objective is to increase the longevity of the network by using dynamic routing. Sensors are energy constrained and are not super powerful like common computers. Every time a sensor transmits some packets, it will decrease the amount of energy left. The team will create a dynamic routing scheme to try and maintain all nodes of the network online for as long as possible.
(3) Hybrid dynamic routing mechanism design
Figure 4 ‑ Network of drones, energy AND throughput are points of concern
Imagine if some drones are collecting some data and sending it to a command and control base (Figure 4). If the application at each drone has low data rate requirement, it does not require too much throughput, instead they could be better off having longer delays in the end-to-end transmission, and saving energy at the same time (Figure 4). For high data rate requirement, the focus is more on throughput. In this project the team will simulate such scenario, combining both throughput and energy. Given that each node only requires a certain amount of throughput, how can the packets be routed so that the network of drones can stay “alive” longer, and the communication to the C&C is not broken?
(4) Dynamic routing mechanism design in faulty network
F
igure 5 ‑ Faulty network: how to route if a node suddenly stops working for a while?
The team will simulate a mesh network where nodes and links may fail (Figure 5). Nodes and links may fail intermittently, as an input to the simulation, each node and link will have a certain probability to fail. When such failure occurs, the network must adapt and re-route to avoid the faulty link/node.
