Amazon Technologies, Inc.

Koohyun Um

Jungtao Liu

Gur Kimchi

Daniel Buchmueller

Subram Narasimhan

Jin Dong Kim

Ronald Joseph Degges, Jr

Sang Eun Lee

In Chul Hyun

Tzung-I Lee

Morris Yuanhsiang Hsu

Feng Zhang

Pratik Kalpesh Patel

Shao-Cheng Wang

Peruvemba Ranganathan Sai Ananthanarayanan

Rongsheng Huang

James J Wolfe

Durgaprasad Kashinath Shamain

Mercyma Deeba Balachandran

A method and system including a monitoring device to detect, during a session, a set of radar signals associated with first respiratory-related movements of a first person and second respiratory-related movements of a second person located within a detection zone comprising at least a portion of a bed. The method and system dynamically generate an out-of-bed zone adjacent to a side of the bed. The out-of-bed zone is monitored to detect motion within the zone to determine when the first person exits and enters the bed. The method and system identify a portion of the set of radar signals collected during a time period when the first person is determined to be out of the bed. The identified portion of the set of radar signals is removed from the set of radar signals collected during the sleep session.

Out-of-bed zone generation and monitoring associated with respiratory activity tracking

Devices, systems, and methods are provided for disambiguating phase in radar-based waveforms. A method may include determining a first waveform, the first waveform associated with a first subcarrier and a second subcarrier in a frequency domain, and the first waveform based on radar measurements. The method may include determining a motion signal, and determining, based on the first subcarrier, the second subcarrier, a subcarrier spacing of the subcarriers, and the motion signal, a phase change differential value between the first subcarrier and the second subcarrier. The method may include generating, based on the phase change differential value, a second waveform, and generating a third waveform in the time domain by applying the second waveform to the first waveform.

Phase disambiguation for radar-based breathing measurements

A system that can determine states of human activity and transitions between those states using wireless signal data. A machine learning model such as a Hidden Markov Model (HMM) may be trained to determine transitions between states of human activity (e.g., static, slow movement, fast movement) using information from wireless signal data, such as channel state information gathered from Wi-Fi signal beacons. Depending on the state of the human activity the system may then cause certain commands to be executed corresponding to the human activity such as turning on a certain configuration of lights, playing certain music, or the like.

Human activity and transition detection

Systems, methods, and apparatus are provided for enabling orientation of directional antennas even when one or more of the directional antennas are moving. Position information for each directional antenna is transmitted using an omnidirectional antenna transmitting at a low bandwidth and a low power. The position information of the directional antennas is used to orient the directional antennas so that a high bandwidth, low power wireless connection can be enabled and/or maintained between the directional antennas. The position information is periodically transmitted and the orientation of the directional antennas is updated as one or more of the directional antennas move so that the wireless connection between the directional antennas is maintained.

Orientation of directional antennas using horizontal position information

Technology for indoor location estimation of wireless stations is described. In one embodiment, a processing device identifies a plurality of wireless stations that support a signal round-trip-time measurement protocol. The processing device associates the plurality of wireless stations with one another to form a logical grouping of the plurality of wireless stations and provides an indication of the plurality of wireless stations in the logical grouping to each of the plurality of wireless stations. The processing device receives a first round-trip-time measurement value for a signal transmitted between a first wireless station and a second wireless station of the logical grouping and a second round-trip-time measurement value for a signal transmitted between the first wireless station and a third wireless station of the logical grouping. These round-trip-time measurement values indicate a distance between the wireless stations, respectively. The processing device determines, using the first distance and the second distance, a position of the first wireless station relative to the second wireless station and the third wireless station.

Indoor location estimation for wireless device

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.

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

A system and method include processing logic receiving, from a wireless transceiver of a first device, first data indicative of channel state information (CSI) of a first communication link between the wireless transceiver and a wireless transmitter of a second device, the first device and the second device being located in a building. The logic pre-preprocesses the first data to generate input vectors composed of statistical parameter values derived from sets of discrete samples of the first data. The logic processes, through a long short-term memory (LSTM) layer of a neural network, the input vectors to generate multiple hidden state values of the neural network. The logic processes, through a set of additional layers of the neural network, respective hidden state values of the multiple hidden state values to determine that a human is present in the building.

Coexistence-insensitive presence detection

Antenna structures and methods of operating the same of an electronic device are described. One apparatus includes an antenna element that is self-resonant at a first frequency in a first mode and self-resonant at a second frequency in a second mode. A radio frequency (RF) is coupled between the antenna element and a ground plane. A RF circuit, coupled to the antenna element via an RF feed, is operable to control the RF switch to switch the antenna element between the first mode and the second mode. The RF circuit is operable to cause the antenna element to radiate or receive electromagnetic energy in a first radiation pattern in the first mode and to radiate or electromagnetic energy in a second radiation pattern in the second mode.

Growing community of inventors

Sunnyvale, CA, United States of America

Koohyun Um