US 10,310,079 B1Granted

Presence detection using wireless signals confirmed with ultrasound and/or audio

AssigneeAmazon Technologies, Inc.

InventorsKoohyun Um, Ce Zhang, Jungtao Liu

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method includes receiving, by a first wireless device, first data indicative of channel properties of a first communication link between the first wireless device and a second wireless device, the first wireless device and the second wireless device being located in a same vicinity. The method further includes detecting, using a supervised machine learning (ML) model applied to the first data, human movement and presence of stationary objects within the same vicinity and activating, by the first wireless device in response to detection of the human movement, an ultrasonic signal. The method further includes receiving, by the first wireless device, a reflected component of the ultrasonic signal and confirming, by the first wireless device using the reflected component, presence of a human in the same vicinity.

Claims20

Dependent claims

A wireless device located at a first location within a building, the wireless device comprising:

an antenna;
a wireless local area network (WLAN) receiver coupled to the antenna;
a communication interface coupled to the WLAN receiver, the communication interface to receive first data, wherein the first data includes channel properties of a first communication link between the WLAN receiver and a WLAN transmitter located at a second location in the building, the channel properties comprising channel state information (CSI);
and
a processor coupled to the communication interface, the processor to:
pre-process the first data to generate pre-processed first data;
execute a fast Fourier transform (FFT) on the pre-processed first data to generate a data stream that contains frequency domain information of the pre-processed first data;
extract frequency components from the data stream to generate Doppler spectrum data, wherein the frequency components comprise data points located between direct current (DC) and approximately 30 Hz, which are indicative of movement of a human within the building;
classify, via execution of a supervised machine learning (ML) model, the Doppler spectrum data to identify (i) the movement of the human within the building and (ii) stationary objects within the building, to detect presence of the human at the first location within the building;
activate, in response to detection of the presence of the human, an ultrasonic signal or an audio signal;
receive a reflected component of the ultrasonic signal or the audio signal;
confirm the presence of the human at the first location based on the reflected component;
and
trigger a light switch to a turn on a light at the first location.

11 more elements

A method comprising:

receiving, by a first wireless device, first data indicative of channel properties of a first communication link between the first wireless device and a second wireless device, the first wireless device and the second wireless device being located in a same vicinity, and the channel properties comprising channel state information (CSI);
detecting, using a supervised machine learning (ML) model applied to the first data, human movement and presence of stationary objects within the same vicinity;
activating, by the first wireless device in response to detection of the human movement, an ultrasonic signal;
receiving, by the first wireless device, a reflected component of the ultrasonic signal;
and
confirming, by the first wireless device using the reflected component, presence of a human in the same vicinity;
and
outputting, by the first wireless device, a signal indicative of confirmed presence of the human in the same vicinity.

5 more elements

A non-transitory computer-readable medium comprising instructions which, when executed by a first wireless device located at a first location within a building, causes the first wireless device to perform operations comprising:

retrieving first data indicative of channel properties of a first communication link between the first wireless device and a second wireless device, the second wireless device being located at a second location in a building, and wherein the channel properties comprise channel state information (CSI);
detecting, via execution of a supervised machine learning (ML) model using the first data, human movement and stationary objects within wireless range of the first location in the building;
triggering activation of an ultrasonic signal in response to detection of the human movement;
retrieving data containing a reflected component of the ultrasonic signal that is received by the first wireless device;
and
confirming, using the data containing the reflected component, presence of a human at the first location;
and
triggering out of a signal indicative of confirmed presence of the human at the first location.

5 more elements

Dependencies are inferred from the claim text.