Delay-tolerant networking effectively extends the network connectivity in the timeuddomain, and endows communications devices with enhanced data transferudcapabilities. Network coding on the other hand enables us to approach theudinformation capacity of networks by allowing intermediate nodes to process data enudroute. Both of these were major principal breakthroughs in mobile and wirelessudcommunications in the past decade or so. In the first half of this thesis, we considerudthe problem of disseminating a large number of messages in such networks. Withudthe sparse and intermittently connected topology and with the unreliable andudlow-rate radios, the strategy of which messages to transfer first and in what order isuda determinant of performance here. We compare a few such message prioritizationudmethods using computer simulation and observe their performance in terms howudwidely and quickly information can be distributed across the network. Next, we areudinterested in how network coding stacks against conventional epidemic routingudvariants. We conducted tests with both real smart mobile devices and computerudsimulation and found conditions where their results match. This would give usudconfidence of using computer simulation to study larger delay-tolerant networksudwith and without network coding at a much manageable cost.
展开▼
