Method and system for dynamic meeting detection using multiple physical features from mobile devices
AssigneePalo Alto Research Center Incorporated
InventorsOliver Brdiczka, Jens Weppner
Life of this patent
Estimating the term…
Abstract
One embodiment of the present invention provides a system for detecting in-person meetings between users of mobile devices. During operation, the system collects information regarding characteristics of features of the users' surrounding environment as detected by the mobile devices. The characteristics of the detected features are compared with each other to produce observations and the observations are applied to a hidden Markov model to detect meetings between the users. Features detected from the surroundings may include Wi-Fi signals, discoverable Bluetooth signals, and ambient sounds.
Claims22
A computer-executable method for detecting in-person meetings between users, the method comprising:
- obtaining, by a server, feature information representing characteristics of a surrounding environment detected by two mobile devices associated with the users, and that indicates a location and/or distance between the two mobile devices, wherein the characteristics include Wi-Fi signals detected by the two mobile devices;
- producing, by the server, input observations to a hidden Markov model from the collected characteristics information, which involves comparing the Wi-Fi signals detected by the two mobile devices to produce the input observation associated with Wi-Fi signals;
- and
- determining, by the server, the presence of an in-person meeting between the users, which involves applying the hidden Markov model comprising a plurality of states that each reflect a meeting state, wherein the hidden Markov model includes at least a state indicating that the users are not in a meeting, and a state indicating that the users are in a meeting.
A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for detecting in-person meetings between users, the method comprising:
- obtaining feature information representing characteristics of a surrounding environment detected by two mobile devices associated with the users, and that indicates a location and/or distance between the two mobile devices, wherein the characteristics include Wi-Fi signals detected by the two mobile devices;
- producing input observations to a hidden Markov model from the collected characteristics information, which involves comparing the Wi-Fi signals detected by the two mobile devices to produce the input observation associated with Wi-Fi signals;
- and
- determining the presence of an in-person meeting between the users, which involves applying the hidden Markov model comprising a plurality of states that each reflect a meeting state, wherein the hidden Markov model includes at least a state indicating that the users are not in a meeting, and a state indicating that the users are in a meeting.
A computing system for detecting in-person meetings between users, the system comprising:
- one or more processors,
- a non-transitory computer-readable medium coupled to the one or more processors having instructions stored thereon that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
- obtaining feature information representing characteristics of a surrounding environment detected by two mobile devices associated with the users, and that indicates a location and/or distance between the two mobile devices, wherein the characteristics include Wi-Fi signals detected by the two mobile devices;
- producing input observations to a hidden Markov model from the collected characteristics information, which involves comparing the Wi-Fi signals detected by the two mobile devices to produce the input observation associated with Wi-Fi signals;
- and
- determining the presence of an in-person meeting between the users, which involves applying the hidden Markov model comprising a plurality of states that each reflect a meeting state, wherein the hidden Markov model includes at least a state indicating that the users are not in a meeting, and a state indicating that the users are in a meeting.
Dependencies are inferred from the claim text.