Date of Graduation
Statler College of Engineering and Mineral Resources
Lane Department of Computer Science and Electrical Engineering
Duty cycle is important in wireless sensor and mobile ad-hoc networks (MANETs) to extend their lifetime. Duty cycling in wireless networks lets the nodes wake up with infrequent fixed periods, based on pre-determined parameters. On the other hand, neighbor discovery is the process by which nodes learn about the neighbors. Nodes can use radio communications to discover other neighbors. Timely, neighbor discovery is crucial for routing. But, neighbor discovery becomes more challenging in energy-constrained, mobile environment with duty cycled nodes where a node may not know whether any neighbors are present, and what duty cycle those neighbors might operate at. Disco is an asynchronous neighbor discovery and rendezvous protocol that allows two or more nodes to operate their radios at low duty cycles (1-2%). Disco obtains discovery and communication during infrequent, opportunistic encounters without requiring a global synchronization information. Disco nodes pick a pair of dissimilar prime number such that the sum of their reciprocal is equal to the desired duty cycle. A global counter increments with a fixed period. If one of the node's prime numbers is divisible by the counter, the node will turn on its radio for one period. This protocol ensures that two nodes have some overlapping radio on-time within the boundary of the period, and discover each other despite of the independent set of duty cycle. In this report, we seek to understand how Disco performs in mobile networks. We analyze the impact of different duty cycles, mobility speeds and network sizes on discovery latency in extreme mobile networks. We also scrutinize how fast Disco discovers that nodes have moved out of their neighborhood. We use NS3 to simulate Disco with different, duty cycles, mobility speeds, and network sizes.
Alramzi, Saud, "Analyzing discovery latency in mobile networks" (2017). Graduate Theses, Dissertations, and Problem Reports. 3995.