Date of Award
Open Access Dissertation
Computer Science and Engineering
In IP backbone networks, packets may get dropped due to: i) lack of viable next hops when a link/router fails, ii) forwarding loops during network convergence, and iii) buffer overflows in case of congestion. Similarly, packets may be lost in wireless networks due to variations in signal strength between a pair of mobile nodes. This dissertation explores the possibility of providing a predictable performance during such network contingencies in wired backbone networks and robotic wireless networks.
First, we study the feasibility of developing a combination of local reroute and global update mechanisms that can achieve loop-free convergence, while performing disruption-free forwarding around a failed link/router, without carrying any additional information in the IP datagrams and with out needing any coordination between routers. We show that order of updates rarely matters for loop-free convergence when failure inference based fast reroute (FIFR) scheme with interface-specific forwarding is employed for dealing with link or router failures. In the rare cases where order matters, it can be coupled with progressive link metric increments to ensure loop-freedom with unordered updates of forwarding tables. We also demonstrate that, apart from providing protection against failures, FIFR can also be utilized to mitigate packet drops due to network congestion caused by micro traffic bursts.
Second, we address the problem of constructing a communication map, which encodes information on whether two robots at given locations can communicate using a wireless network. Unlike previous offline approaches that do not utilize data measured by robots, we propose an online method, utilizing Gaussian Processes, to efficiently build a communication map with multiple robots, by exploiting prior communication models that can be derived from the physical map of the environment. Our evaluation, using a team of TurtleBot 2 platforms, confirms that the proposed method requires robots to take fewer signal strength measurements and travel less distance, and yet obtain similar accuracy as methods that consider all the locations in the environment.
Penumarthi, P. K.(2020). Providing Predictable Performance during Network Contingencies. (Doctoral dissertation). Retrieved from https://scholarcommons.sc.edu/etd/6171