Characterization of Random Linear Network Coding with Application to Broadcast Optimization in Intermittently Connected Networks

We address the problem of optimizing the throughput of network coded traffic in mobile networks operating in challenging environments where connectivity is intermittent and locally available memory space is limited. Random linear network coding (RLNC) is shown to be equivalent (across all possible initial conditions) to a random message selection strategy where nodes are able to exchange buffer occupancy information during contacts. This result creates the premises for a tractable analysis of RLNC packet spread, which is in turn used for enhancing its throughput under broadcast. By exploiting the similarity between channel coding and RLNC in intermittently connected networks, we show that quite surprisingly, network coding, when not used properly, is still significantly underutilizing network resources. We propose an enhanced forwarding protocol that increases considerably the throughput for practical cases, with negligible additional delay.