Discovering hotspots: A placement strategy for Wi-Fi based trajectory monitoring within buildings

摘要

In the last decade, Wi-Fi based trajectory monitoring has gathered high interest in the scientific and commercial world, due to the increased usage of Wi-Fi capable mobile devices. A lot of work can be found where monitor nodes are placed in an area of interest capturing Wi-Fi signals from passing phones. However, the deployment of such nodes is often inefficient, expressed by a low ratio between monitored trajectories and the amount of installed nodes. Hence, finding an optimal setting of node positions is an essential and challenging task. In this paper, a systematic solution for this variant of the NP-hard art gallery problem is investigated. The idea is to set monitor nodes only on places (hotspots), where most of the human paths can be tracked. For the discovery of such hotspots, three novel approaches are presented working on simulated user traces based on an extended pathway mobility model, and a given plant layout. The results of each approach are evaluated in terms of quality for Wi-Fi based trajectory monitoring using different parameters and settings. The evaluation indicates that the proposed methods show different potentials and limitations. Overall, they return a reliable setting of hotspots compared to a completely random selection of an equal amount of node positions and, thus, they serve as a systematic and sophisticated placement strategy.