Systems and methods for room-level location using WiFi

摘要

A method of identifying a location of a mobile device in a building includes identifying non-overlapping regions in a building. A server collects base station signal strength measurements at a plurality of distinct points in the building, with at least one point in each region. The server trains region classifiers for each region. Each region classifier is configured to compute a probability estimate that the test point is inside the region, using inputs that are signal strength differences. The server receives signal strength measurements from the base stations, taken by a mobile device at an unknown point. The server computes differences in signal strengths between pairs of base stations, and applies the region classifiers to the signal strength differences, thereby estimating the region where the mobile device is located. The server then transmits the estimated region to a user.

主权项

1. A method of identifying locations of mobile devices in buildings, performed at a server having one or more processors and memory, the method comprising:
partitioning a building into a plurality of regions, wherein the building includes a plurality of radio-transmitting base stations at distinct locations; receiving training data measurements taken at a plurality of distinct points in the building, wherein for each of the distinct points the measurements measure strength of signals received from the base stations, and wherein at least one respective point of the distinct points is in each of the respective regions; for each respective point of the plurality of distinct points in the building, computing differences between pairs of signal strength measurements taken at the respective point and forming a respective feature vector that includes a plurality of components, each component comprising a respective computed difference associated with a respective base station pair, thereby forming a plurality of feature vectors; and for each region, using the feature vectors to train a respective region classifier, wherein the training includes selecting a respective subset of the base station pairs that consists of fewer than all base station pairs and determining a respective plurality of parameters corresponding to the respective subset of base station pairs, and wherein the respective plurality of parameters are modified during the training so that the respective region classifier computes probability estimates that test points are inside the respective region using sample feature vectors of signal strength differences for signal strength measurements taken at the test points for signals received from the base stations.