Electronic device locations inferences
AssigneeHewlett-Packard Development Company, L.P.
InventorsPei Hsuan Li, Janani Anand, Christopher Charles Mohrman
Life of this patent
Estimating the term…
Abstract
In some examples, an electronic device includes a wireless communication device. In some examples, the electronic device includes a transceiver of the wireless communication device to receive radio frequency (RF) signals. In some examples, the electronic device includes a processor. In some examples, the processor is to generate a data cluster based on the RF signals. In some examples, the data cluster corresponds to a location of the electronic device. In some examples, the processor is to determine a classifier based on the cluster. In some examples, the processor is to infer, via a machine learning model, the location of the electronic device based on the classifier.
Claims14
An electronic device, comprising:
- a wireless communication device;
- a transceiver of the wireless communication device to receive radio frequency (RF) signals;
- and
- a processor to:
- generate a data cluster based on the RF signals, wherein the data cluster corresponds to a location of the electronic device;
- discard an identifier of an RF signal that is observed for less than a threshold period;
- determine a classifier based on the data cluster;
- and
- infer, via a machine learning model, the location of the electronic device based on the classifier.
An apparatus, comprising:
- a wireless local area network (WLAN) transceiver to receive a WLAN identifier;
- and
- a processor to:
- add the WLAN identifier to an identifier set in response to determining that the WLAN identifier has been received for a threshold period;
- determine a fingerprint vector based on the identifier set;
- generate a data cluster based on the fingerprint vector;
- train a classifier based on the data cluster;
- and
- add the classifier to an ensemble model to infer a location corresponding to the data cluster.
A non-transitory tangible computer-readable medium comprising instructions when executed cause a processor of an electronic device to:
- assign a detection indicator for an identified wireless signal source in an identifier set;
- determine, using a first machine learning model, a data cluster based on the identifier set, wherein the data cluster corresponds to a location;
- train a one-class support vector machine (SVM) classifier based on the data cluster;
- and
- infer, using a second machine learning model that includes the one-class SVM classifier, the location based on a received wireless signal.
Dependencies are inferred from the claim text.