Inferring Wireless Communications Links and Network Topology from Externals using Granger Causality

This paper presents Granger Causality as a method for inferring communications links among a collection of wireless transmitters from externally measurable features. The link inference method presented relies upon general assumptions that hold true for a wide variety of communications, and is therefore applicable to inferring the link topology of broad classes of wireless networks, regardless of the nature of the Medium Access Control (MAC) protocol used. This technique does not require decoding of data and can be used to infer links based upon features of communications observable from outside the network. We illustrate the use of this method on simulated NS3 data to infer the topology of ad-hoc 802.11 networks. The accuracy, convergence rate, and robustness to noise of link inference are presented for networks of different sizes, link densities, etc.