Abstract: Methods and apparatus are disclosed herein for computation and compression efficiency in distributed video analytics. Example apparatus disclosed herein are to identify a key frame and a non-key frame in a video frame sequence input to a neural network at a client server, determine motion information between the key frame and the non-key frame based on optical flow, and determine a frame feature representation based on the motion information reconstructed at an edge server, the motion information including feature warping residual errors.

Abstract: The present disclosure is related to power control mechanisms for workload processing systems, and in particular, multi-scale power control technologies that can be used to reduce the overhead of workload processing systems. The disclosed power control mechanisms operate on multiple timescales including a slow timescale and a fast timescale. Separate control loops (or governors) are used for the slow and fast timescales where each control loop includes its own trigger mechanisms and configurable operational policies. The operational policies for slow timescale control loop can be trained separately using various machine learning techniques while the operational policies for the fast timescale control loop can be simple and reactive heuristics.

Abstract: Various systems and methods for route-based digital service management are described herein, comprising receiving a request for service, such as a user request for service from a MaaS or digital service provider, estimating digital service usage with respect to the request, determining a MaaS route using the request and the estimated digital service usage, selecting an orchestration strategy comprising a server type using the estimated digital service usage, and scheduling a MaaS vehicle for the determined MaaS route in response to the request using the selected orchestration strategy and the determined candidate MaaS route.

Abstract: Systems and methods may be used to map and collect data from a mesh network at a gateway device connecting a plurality of devices (e.g., edge devices, IoT devices, sensors, or the like) to a network. A method may include determining a shortest path tree (SPT) map of the plurality of devices, the shortest path tree map may define the mesh network for the plurality of devices connected to the gateway. The method may include sampling data throughout the mesh network based on a compressive sensing (CS) sampling schedule. The sampled data may be output, such as to a remote device via the network. The sampled data may be saved at the gateway.

Abstract: Techniques disclosed for accurately predicting the occurrence of anomalous sensor readings within a sensor network and advantageously using these predictions to limit the amount of power used by relay nodes within the sensor network. Some examples analyze spatial and temporal characteristics of anomalous sensor readings to predict future occurrences. In these examples, the relay nodes operate in a reduced power mode for periods of time in which anomalous sensor readings are not predicted to occur. Also, in these examples, only relay nodes in a path between a sensor predicting an anomalous reading and a gateway of the sensor network operate in full power mode. This feature allows other relay nodes to remain in the reduced power mode even when an anomalous sensor reading is predicted elsewhere in the sensor network.

Abstract: Aspects of traffic aware slot assignment are described, for example, in a multi-hop wireless network comprising a plurality of nodes. In some aspects, an apparatus of a wireless device is configured to decode signaling, received from a node of the multi-hop network, to determine an indication of a change to a topology of the multi-hop network. The apparatus is further configured to, in response to a determination, from the decoded signaling, of an addition of a second node to the multi-hop network topology, increment a total of a number of descendant nodes, and allocate one or more transmission slots to a number of unused slots in one or more transmission opportunity regions of a slotframe, wherein the slotframe includes a repeating pattern of one or more transmission opportunity periods for a plurality of nodes in the network.

Abstract: Apparatuses, systems and methods associated with adaptive communication topology within a wireless mesh network of devices are disclosed herein. In embodiments, a computer device, may include wireless circuitry to communicate with a leader of a network and a controller coupled to the wireless circuitry. The controller may identify one or more values located within a message received via the network, wherein the one or more values are generated by the leader, and determine in which of one or more operation modes the computer device is to operate based, at least partially, on the one or more values. Other embodiments may be described and/or claimed.

Abstract: Aspects of traffic aware slot assignment are described, for example, in a multi-hop wireless network comprising a plurality of nodes. In some aspects, an apparatus of a wireless device is configured to decode signaling, received from a node of the multi-hop network, to determine an indication of a change to a topology of the multi-hop network. The apparatus is further configured to, in response to a determination, from the decoded signaling, of an addition of a second node to the multi-hop network topology, increment a total of a number of descendant nodes, and allocate one or more transmission slots to a number of unused slots in one or more transmission opportunity regions of a slotframe, wherein the slotframe includes a repeating pattern of one or more transmission opportunity periods for a plurality of nodes in the network.

Abstract: Systems and methods may be used to map and collect data from a mesh network at a gateway device connecting a plurality of devices (e.g., edge devices, IoT devices, sensors, or the like) to a network. A method may include determining a shortest path tree (SPT) map of the plurality of devices, the shortest path tree map may define the mesh network for the plurality of devices connected to the gateway. The method may include sampling data throughout the mesh network based on a compressive sensing (CS) sampling schedule. The sampled data may be output, such as to a remote device via the network. The sampled data may be saved at the gateway.

Abstract: Techniques disclosed for accurately predicting the occurrence of anomalous sensor readings within a sensor network and advantageously using these predictions to limit the amount of power used by relay nodes within the sensor network. Some examples analyze spatial and temporal characteristics of anomalous sensor readings to predict future occurrences. In these examples, the relay nodes operate in a reduced power mode for periods of time in which anomalous sensor readings are not predicted to occur. Also, in these examples, only relay nodes in a path between a sensor predicting an anomalous reading and a gateway of the sensor network operate in full power mode. This feature allows other relay nodes to remain in the reduced power mode even when an anomalous sensor reading is predicted elsewhere in the sensor network.

Abstract: Based upon the present location and historical usage at that location, the technologies disclosed herein select a wireless peripheral device (e.g., a printer, monitor, keyboard) for wireless connection with a portable device (e.g., a laptop computer or tablet). This Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Abstract: A technology is described for determining a position of a HBC (Human Body Channel) sensor. An example method may include receiving signal data for a radio signal from a first transmitting HBC sensor transmitted over a human body channel. A signal loss of the radio signal can be calculated using the signal data received from the first transmitting HBC sensor, where the signal loss may be a function of distance of the first transmitting HBC sensor from a receiver. A distance of the first transmitting HBC sensor from the receiver can then be determined based in part on the signal loss and a relative position of the first transmitting HBC sensor can be identified based in part on the distance between the first transmitting HBC sensor and the receiver.

Abstract: A technology is described for determining a position of a HBC (Human Body Channel) sensor. An example method may include receiving signal data for a radio signal from a first transmitting HBC sensor transmitted over a human body channel. A signal loss of the radio signal can be calculated using the signal data received from the first transmitting HBC sensor, where the signal loss may be a function of distance of the first transmitting HBC sensor from a receiver. A distance of the first transmitting HBC sensor from the receiver can then be determined based in part on the signal loss and a relative position of the first transmitting HBC sensor can be identified based in part on the distance between the first transmitting HBC sensor and the receiver.

Abstract: Apparatuses, systems and methods associated with adaptive communication topology within a wireless mesh network of devices are disclosed herein. In embodiments, a computer device, may include wireless circuitry to communicate with a leader of a network and a controller coupled to the wireless circuitry. The controller may identify one or more values located within a message received via the network, wherein the one or more values are generated by the leader, and determine in which of one or more operation modes the computer device is to operate based, at least partially, on the one or more values. Other embodiments may be described and/or claimed.

Abstract: Data can be transferred from one device to another in the Internet of Things without using a network by a touch-based human body communication (HBC) interaction between a wearable storage module and HBC-compatible interface pads on external host devices. Information on a source host device is copied to the wearable storage module when the user touches the source device's HBC interface pad, can be stored indefinitely on the wearable module, and is copied to a destination host device when the user touches the destination devices HBC interface pad. Because the interface pads only need to be simple electrodes, their size and shape can be widely varied to fit the host devices.

Abstract: Communication devices and techniques for facilitating dual-mode communication within a single wireless communication device are described. In one embodiment, for example, an apparatus may include at least one memory and logic for a wireless communication device, at least a portion of the logic comprised in hardware coupled to the at least one memory, the logic to perform a transmission determination to determine whether to transmit a signal in a low-energy mode or a body-communication mode, map the signal to a body-communication channel responsive to the transmission determination indicating the body-communication mode, and map the signal to a low-energy channel responsive to the transmission determination indicating the low-energy mode. Other embodiments are described and claimed.

Abstract: Systems, apparatuses, and methods may include a human body communication data storage device having at least first and second electrodes and a human body communication modem. A storage component communicating with the human body communication modem includes a first secure storage location provided with a user-specific authentication record and a second data storage location.

Abstract: In various embodiments, two wireless communication devices may communicate with each other using multiple protocols, by dividing the data to be communicated into multiple portions, and using each protocol to communicate different portions. The different protocols may be used simultaneously or concurrently. This multi-protocol technique may be used in several different ways to provide different types of advantages in wireless communications.

Abstract: Described herein are technologies related to estimating location of a mobile device especially while the device is traveling a known and mapped route. That is, the described technologies estimate a user's location when they are traversing a commonly traveled route. More particularly, the described technologies are especially suited to estimating geo-location of a user. This Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Abstract: A garment includes a passive human body communication (HBC) component that includes, for example, a storage element. The garment has conductive cuffs and a flexible conductive trace connecting the cuffs to the HBC component. When a user wearing the garment touches the electrodes of an HBC interface on an external host device, the host device powers the HBC component and may send or receive data from the HBC component. The power and the data travel over the user's body from the interface electrodes to the cuffs, and at least partially through the conductive trace from the cuffs to the HBC component.