Comcast Cable Communications, LLC

Tianwen Chen

Hongcheng Wang

Hong Li

Yonatan Vaizman

Ding Zhou

At least one device can be classified based on user activity. At least one telemetry variable with respect to time may be received for the at least one device. The at least one telemetry variable with respect to time may be used to determine an activity model for the at least one device. Based on the activity model, a probability of the at least one device belonging to a profile or to a type of user may be determined. Based on the type of user of the at least one device, a setting associated with the at least one device may be changed.

Analyzing user activity

Systems, apparatuses, and methods are described for presence detection of a wireless device in a sensing region. Communication parameters determined by a computing device over one or more periods may be used to generate criteria for determining the occurrence of enter events or leave events. The criteria may comprise, for values of one or more communication parameters, rates of change associated with movement into or out of the sensing region. The criteria may be compared against subsequent data associated with the communication parameters to determine if an enter event or a leave event has occurred.

Wireless based presence detection

Monitoring a premises may comprise receiving signals from a plurality of devices. One or more spatially static devices of the plurality of devices may be determined based on strengths of the received signals. A user activity pattern may be determined based on changes in strengths of signals received from the spatially static devices. Abnormal user activity may be determined based on the user activity pattern and a strength of at least one signal received from at least one of the spatially static devices.

Systems and methods for premises monitoring

Methods, systems, and apparatuses for managing a wireless network are described herein. A wireless network may include a gateway device, an access point, and a group of client devices. An analytics engine may determine a signal characteristic data for a client device, which may indicate network telemetry data for each of the client devices or associated locations during a time period. The analytics engine may utilize a trained classifier to determine that a signal characteristic indicated by the signal characteristic data for a client device satisfies a threshold.

Methods, systems, and apparatuses for presence detection

Systems, apparatuses, and methods are described for presence detection of wireless device in a sensing region. Communication parameters determined by a computing device over one or more periods may be used to generate criteria for determining occurrence of enter events or of leave events. The criteria may comprise, for values of one or more communication parameters, rates of change associated with movement into or out of a sensing region. The criteria may be compared against subsequent data, associated with the communication parameters, to determine if an enter event or a leave event has occurred.

Growing community of inventors

McLean, VA, United States of America

Tianwen Chen