On the Estimation of Spatial Density From Mobile Network Operator Data

We tackle the problem of estimating spatial distribution mobile phones from Mobile Network Operator (MNO) data, namely Call Detail Record (CDR) or signalling data. The process transforming MNO data to a density map requires geolocating radio cells determine their footprint. Traditional geolocation solutions rely on Voronoi tes sellations and approximate cell footprints by mutually disjoint regions. Recently, some pioneering work started consider more elaborate methods with partially overlapping (non- disjoint) coupled probabilistic model for phone-to-cell association. Estimating in such setup is currently an open research focus present work. start reviewing three different estimation proposed literature provide novel analytical insights that unveil key aspects mutual relationships properties. Furthermore, we develop approach which closed-form solution can be given. Numerical results based semi-synthetic are presented assess relative accuracy each method. Our indicate estimators have potential improve over traditional approaches tessellations.