System and method for classifying a type of interaction between a human user and a mobile communication device in a volume based on sensor fusion

A system for classifying a type of interaction between a human user and at least one mobile communication device within a defined volume, based on at least one sensor, the system comprising:

• a phone locating unit configured to determine a position of the at least one mobile communication device relative to a frame of reference of the defined volume, based on at least one of:
• angle of arrival, time of flight, or received intensity of radio frequency (RF) signals transmitted by the at least one mobile communication device and received by the phone location unit located within the defined volume and comprising a transceiver and antennas and further configured to wirelessly communicate with the at least one mobile communication device;
• a plurality of non-RF sensors configured to repeatedly obtain at least one sensor measurement related to the at least one mobile communication device, from a sensor located on at least one of:
• the at least one mobile communication device, within the defined volume, or outside of the defined volume, to yield at least one time series of sensor readings, wherein the at least one sensor measurement related to the at least one mobile communication device is obtained from a proximity sensor, said proximity sensor providing information on a proximity between the at least one mobile communication device and at least one human user;
• and
• a computer processor configured to:
• increase a position accuracy of the at least one mobile communication device relative to the frame of reference based on the at least one sensor measurement related to the at least one mobile communication device, to yield an improved position;
• and
• classify the at least one type of interaction into at least one of a plurality of predefined types of interaction, based on the improved position and the at least one time series of sensor readings, wherein the proximity is factored in classifying the likelihood of the type of user-interactions, wherein a closer proximity infers a higher probability of an audio engagement type of interaction.

A non-transitory computer-readable medium for classifying a type of interaction between a human user and at least one mobile communication device within a defined volume, based on at least one sensor, the computer-readable medium comprising a set of instructions that when executed cause at least one computer processor to:

• determine a position of the at least one mobile communication device relative to a frame of reference of the defined volume, based on at least one of:
• angle of arrival, time of flight, or received intensity of radio frequency (RF) signals transmitted by the at least one mobile communication device and received by a phone location unit located within the defined volume and comprising a transceiver and antennas and further configured to wirelessly communicate with the at least one mobile communication device;
• obtain at least one measurement related to the at least one mobile communication device, from a sensor located on at least one of:
• the at least one mobile communication device, within the defined volume, or outside of the defined volume;
• repeat the obtaining over time, to yield at least one time series of the sensor readings;
• and
• increase a position accuracy of the at least one mobile communication device relative to the frame of reference based on the at least one sensor measurement related to the at least one mobile communication device, to yield an improved position, wherein the at least one sensor measurement related to the at least one mobile communication device is obtained from a proximity sensor, said proximity sensor providing information on a proximity between the at least one mobile communication device and at least one human user;
• process the sensor readings and the position of the at least one mobile communication device to yield at least one type of interaction;
• and
• classify the at least one interaction into at least one of a plurality of predefined types of interaction, based on the improved position and the at least one time series of sensor readings, wherein the proximity is factored in classifying the likelihood of the type of user-interactions, wherein a closer proximity infers a higher probability of an audio engagement type of interaction.

A server for classifying a type of interaction of at least one mobile communication device within a defined volume, based on at least one sensor, the server comprising:

• a network interface configured to receive, from a phone location unit, a position of the at least one mobile communication device relative to a frame of reference of the defined volume, based on at least one of:
• angle of arrival, time of flight, or received intensity of radio frequency (RF) signals transmitted by the at least one mobile communication device and received by the phone location unit located within the defined volume and comprising a transceiver and antennas and further configured to wirelessly communicate with the at least one mobile communication device;
• a memory interface configured to store at least one measurement related to the at least one mobile communication device, from a sensor located on at least one of:
• the at least one mobile communication device, within the defined volume, or outside of the defined volume;
• a processing device configured to repeat the obtaining over time, to yield at least one time series of the sensor readings and increase a position accuracy of the at least one mobile communication device relative to the frame of reference based on the at least one sensor measurement related to the at least one mobile communication device, to yield an improved position, wherein the at least one sensor measurement related to the at least one mobile communication device is obtained from a proximity sensor, said proximity sensor providing information on a proximity between the at least one mobile communication device and at least one human user;
• and
• to classify the at least one interaction into at least one of a plurality of predefined types of interaction, based on the improved position and the at least one time series of sensor readings, wherein the proximity is factored in classifying the likelihood of the type of user-interactions, wherein a closer proximity infers a higher probability of an audio engagement type of interaction.