Towards managing age of network state information in challenged networks

Age of Information (AoI) is a recently introduced metric for freshness of the real-time information, which requires frequent updates. This paper argues that AoI is useful not only for real-time network applications, but also for network state information in challenged network management, where conventional TCP/IP bases Internet architectures are not effective due to intermittent connectivity, frequent nodal churn, widely varying network conditions, etc. Since network state updates take bandwidth from the payload carried by the network, we quantify the corresponding tradeoffs and propose an extension to Network Utility Maximization (NUM) framework which incorporates these tradeoffs.