Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals that propagate along two or more paths of a wireless communication channel. The method includes storing a set of eigenvectors derived from first motion-sensing data associated with a first time frame. The first motion-sensing data is associated with a first motion-sensing topology of a wireless mesh network. The method also includes obtaining a motion vector based on wireless signals transmitted between access point nodes in the wireless mesh network during a second, subsequent time frame. The wireless mesh network operates in a second, distinct motion-sensing topology during the second time frame. The motion vector is compared with the respective eigen vectors, and a probability vector is generated based on the comparison. A location of the motion of the object during the second time frame is determined based on the probability vector.

Abstract: In a general aspect, various fields of a PHY frame are used for motion detection. In some aspects, a first training field and a second, different training field are identified in a PHY frame of each wireless signal transmitted between wireless communication devices in a wireless communication network. A first time-domain channel estimate and a second time-domain channel estimate are generated for each wireless signal. The first time-domain channel estimate is based on a first frequency-domain signal included in the first training field, while the second time-domain channel estimate is based on a second frequency-domain signal included in the second training field. A determination is made whether motion has occurred in a space during the time period based on the first time-domain channel estimates, and a location of the motion within the space is determined based on the second time-domain channel estimates.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals and topologies of wireless connectivity. The method includes obtaining motion-sensing data from access point (AP) nodes of a wireless mesh network. The motion-sensing data is based on wireless signals transmitted between respective pairs of the AP nodes. The method additionally includes identifying a motion-sensing topology of the wireless mesh network. The motion-sensing topology is based on tags assigned to respective AP nodes, each tag indicating a connected state of a respective AP node. The method further includes generating a probability vector based on the motion-sensing data and the motion-sensing topology. The probability vector includes values that represent probabilities of motion of an object at respective AP nodes. A location of the motion of the object is determined based on the probability vector.

Abstract: In a general aspect, various fields of a PHY frame are used for motion detection. In some aspects, a first training field and a second, different training field are identified in a PHY frame of each wireless signal transmitted between wireless communication devices in a wireless communication network. A first time-domain channel estimate and a second time-domain channel estimate are generated for each wireless signal. The first time-domain channel estimate is based on a first frequency-domain signal included in the first training field, while the second time-domain channel estimate is based on a second frequency-domain signal included in the second training field. A determination is made whether motion has occurred in a space during the time period based on the first time-domain channel estimates, and a location of the motion within the space is determined based on the second time-domain channel estimates.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals in a wireless mesh network that includes leaf nodes. The method includes obtaining motion-sensing data based on wireless signals exchanged on wireless links in a wireless mesh network including a plurality of nodes. The plurality of nodes includes a first access point (AP) node, one or more other AP nodes, and leaf nodes. The method also includes identifying, based on the motion-sensing data, the first AP node as an estimated location of motion of an object. The method additionally includes generating a likelihood data structure comprising likelihood values assigned to respective nodes of the plurality of nodes in response to the first AP node being identified as the estimated location of motion. A location of the motion of the object is determined based on the likelihood data structure.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals and topologies of wireless connectivity. The method includes obtaining motion-sensing data from access point (AP) nodes of a wireless mesh network. The motion-sensing data is based on wireless signals transmitted between respective pairs of the AP nodes. The method additionally includes identifying a motion-sensing topology of the wireless mesh network. The motion-sensing topology is based on tags assigned to respective AP nodes, each tag indicating a connected state of a respective AP node. The method further includes generating a probability vector based on the motion-sensing data and the motion-sensing topology. The probability vector includes values that represent probabilities of motion of an object at respective AP nodes. A location of the motion of the object is determined based on the probability vector.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals that propagate along two or more paths of a wireless communication channel. The method includes storing a set of eigenvectors derived from first motion-sensing data associated with a first time frame. The first motion-sensing data is associated with a first motion-sensing topology of a wireless mesh network. The method also includes obtaining a motion vector based on wireless signals transmitted between access point nodes in the wireless mesh network during a second, subsequent time frame. The wireless mesh network operates in a second, distinct motion-sensing topology during the second time frame. The motion vector is compared with the respective eigen vectors, and a probability vector is generated based on the comparison. A location of the motion of the object during the second time frame is determined based on the probability vector.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals in a wireless mesh network that includes leaf nodes. The method includes obtaining motion-sensing data based on wireless signals exchanged on wireless links in a wireless mesh network including a plurality of nodes. The plurality of nodes includes a first access point (AP) node, one or more other AP nodes, and leaf nodes. The method also includes identifying, based on the motion-sensing data, the first AP node as an estimated location of motion of an object. The method additionally includes generating a likelihood data structure comprising likelihood values assigned to respective nodes of the plurality of nodes in response to the first AP node being identified as the estimated location of motion. A location of the motion of the object is determined based on the likelihood data structure.

Abstract: In a general aspect, a method is presented for detecting a location of motion using wireless signals and differences between topologies of wireless connectivity. The method includes storing first motion-sensing statistics derived from first motion-sensing data associated with a first time frame. The first motion-sensing data is associated with a first motion-sensing topology of a wireless mesh network. The method also includes obtaining second motion-sensing statistics derived from second motion-sensing data associated with a second, subsequent time frame. The second motion-sensing data is associated with a second, distinct motion-sensing topology of the wireless mesh network. The first motion-sensing statistics are compared to the second motion-sensing statistics. The method additionally includes generating a probability vector based on the comparison. A location of the motion of the object during the second time frame is determined based on the probability vector.

Abstract: In a general aspect, a method is presented for determining a motion zone for a location of motion detected by wireless signals. In some aspects, the method includes obtaining a time series of statistical parameters derived from sets of channel response data. The method also includes identifying time intervals in the time series of statistical parameters, each time interval associated with a respective motion zone in the space. The method additionally includes determining, by analyzing multiple time windows within each of the time intervals, ranges of motion-zone parameters associated with each respective motion zone. The method further includes storing, in a database of a motion detection system, the ranges of motion-zone parameters for the respective motion zones. The ranges of motion-zone parameters are used to identify one of the motion zones based on a motion event detected by the motion detection system.

Abstract: In a general aspect, motion is detected using wireless signals. In an example, a method includes receiving, at a wireless communication device, requests for the wireless communication device to transmit wireless signals, the requests initiated by a wireless sensing system. The method further includes transmitting a series of wireless signals from the wireless communication device in response to the requests, and detecting, at the wireless communication device, a trigger event after transmitting the series of wireless signals. The method additionally includes updating, by the wireless communication device, a state of the wireless communication device based on the trigger event, the updated state indicating that the wireless communication device is not enabled to transmit wireless signals in response to the requests from the wireless sensing system. The method also includes communicating, by the wireless communication device, the updated state of the wireless communication device to the wireless sensing system.

Abstract: In a general aspect, motion is detected using wireless signals. In an example, a method includes receiving first motion detection output data representing a degree of motion detected by a wireless sensing system based on wireless signals exchanged in a wireless communication network. The wireless communication network includes a first access point device and a first client device communicatively coupled by a first wireless communication link. The method also includes receiving network information indicative of a network topology of the wireless communication network, and generating a first control signal configured to change a characteristic of the first wireless communication link based on the first motion detection output data and network information.

Abstract: In a general aspect, motion is detected using wireless signals. In one example, beamforming dynamic information is obtained based on a set of wireless signals transmitted through a space from a first wireless communication device to a second wireless communication device. Motion of an object in the space is detected based on the beamforming dynamic information.

Abstract: In a general aspect, a method is presented for determining a motion zone for a location of motion detected by wireless signals. In some aspects, the method includes obtaining a time series of statistical parameters derived from sets of channel response data. The method also includes identifying time intervals in the time series of statistical parameters, each time interval associated with a respective motion zone in the space. The method additionally includes determining, by analyzing multiple time windows within each of the time intervals, ranges of motion-zone parameters associated with each respective motion zone. The method further includes storing, in a database of a motion detection system, the ranges of motion-zone parameters for the respective motion zones. The ranges of motion-zone parameters are used to identify one of the motion zones based on a motion event detected by the motion detection system.

Abstract: In a general aspect, a method for determining a location of motion detected by wireless communication devices in a wireless communication network includes obtaining motion data associated with a first time frame. The motion data includes a set of motion indicator values. The method also includes generating a first probability vector based on the set of motion indicator values and obtaining a second probability vector generated from motion data associated with a prior time frame. The method additionally includes obtaining a transition probability matrix that includes transition values and non-transition values. The method further includes determining, by operation of a data processing apparatus, a location of the motion detected from the wireless signals exchanged during the first time frame.

Abstract: In a general aspect, a method for determining a location of motion detected by wireless communication devices in a wireless communication network includes obtaining motion data associated with a first time frame. The method also includes identifying a first wireless link, of a plurality of wireless links, based on a magnitude of a motion indicator value associated with the first wireless link relative to the other motion indicator values in a set of motion indicator values. The method additionally includes generating a first probability vector based on a predetermined map and the first wireless link and obtaining a second probability vector generated from motion data associated with a prior time frame. The method further includes determining, by operation of a data processing apparatus, a location of motion detected from wireless signals exchanged during the first time frame.

Abstract: In a general aspect, motion is detected using wireless signals. In one example, a wireless signal transmitted through a space from a first wireless communication device is received at a second wireless communication device. Dynamic beamforming information is based on the wireless signal. The second wireless communication device uses the dynamic beamforming information to detect motion in the space, or transmits the dynamic beamforming information to the first wireless communication device for use in detecting motion in the space.

Abstract: In a general aspect, motion is detected using wireless signals. In one example, a wireless standard protocol message is transmitted by a wireless communication device through a space to one or more neighbor devices. A corresponding wireless standard protocol response including beamforming information feedback is received from each of the one or more neighbor devices in response to the wireless standard protocol message. Motion of an object in the space is detected based on the beamforming information feedback from each of the neighbor devices.

Abstract: In a general aspect, an offline system detects new motion zones for a motion detection system. In some examples, observed channel response data is obtained from a motion detection system. The observed channel response data is associated with a plurality of subcarriers for each wireless link over a period of time. Estimated channel response data is generated for each wireless link. Transitions between zones of activity are identified in the estimated channel response data for each of the plurality of subcarriers for each wireless link. A new motion zone of the motion detection system is detected based on the estimated channel response data and the identified transitions.

Abstract: In a general aspect, a method for determining a location of motion detected by wireless communication devices in a wireless communication network includes obtaining a set of motion indicator values associated with a prior time frame. The method also includes generating a prior probability vector based on the set of motion indicator values and comprising values for wireless communication devices connected to the wireless communication network during the prior time frame. The method additionally includes selecting the prior probability vector or one of a plurality of initialization probability vectors based on the set of motion indicator values, a configuration of the wireless communication network, or both. The method further includes using the selected probability vector and a set of motion indicators associated a second, subsequent time frame to identify a location associated with motion that occurred during the subsequent time frame.