More efficient mobile device location data can be obtained by estimating a most likely location point in a coverage pattern using a kernel density estimation technique. The kernel density estimation technique can provide a continuous estimate of the most frequented locations of a mobile device(s) within a coverage area. For each wireless sector, the collected location data can grouped to the closest geographic coordinate system, and an inference can be made based on the grouped data.