Mobile Crowd-Sensing Wireless Activity with Measured Interference Power

In this paper, we present a novel method for sensing the volume of wireless throughput of other users by measuring the passive interference power received on mobile devices. The benefit of this mobile crowd-sensing approach is that it offers a non-intrusive way of inferring the level of wireless traffic, without requiring to extract data from commercial or private devices. The availability of cross-network wireless traffic data can allow individual operators to deploy femto-cells effectively. The proposed technique requires only approximate location data and is independent of the traffic pattern. The results show that the sensing accuracy for cell-level activity is high (<; 5% error). The paper further demonstrates how this novel method can allow researchers to rapidly build up a map of the wireless activity level in a large region using crowd-sensing.