An Autoregressive Estimator for Overhead Reduction in Substitution Networks

A substitution network is a temporary network that self-deploys to dynamically replace a portion of a damaged infrastructure by means of a fleet of mobile routers. Some efficient solutions deploy robots based on active measurements. A robot/node in the network may use active link monitoring to assess the link quality towards its neighbors through the use of probe packets. Such probe packets are sent periodically at a given rate, and so, the accuracy of the measurements depends on the number and the frequency of exchanged packets. However, exchanging probe packets is energy and bandwidth consuming, thus active monitoring is considered as a costly mechanism. Even so, active link monitoring is a technique widely used on many network protocols. In this paper, we focus on an adaptive positioning algorithm (APOLO) to self-deploy a network. APOLO is based on active monitoring to gather essential information from nodes. Therefore, we show how autoregressive estimation may be used to reduce the overhead caused by the active measuring technique. Moreover, it is possible to use surrogate data rather than real data to feed APOLO without impacting its performance.