Abstract:
In one implementation, a method performed by a client device includes: determining a perceptual characterization metric, where the perceptual characterization metric provides a quantitative characterization of factors that determine quality of experience associated with a first playback profile of media content presented by the client device; and normalizing the perceptual characterization metric based on characterization factors associated with the client device. The method further includes: obtaining an aggregate perceptual characterization metric that is based at least in part on the normalized perceptual characterization metric from the client device and one or more other perceptual characterization metrics from other client devices sharing at least one network resource with the client device; and determining whether to change the first playback profile based on a function of the first normalized perceptual characterization metric and the aggregate perceptual characterization metric.
Abstract:
Presented herein are techniques for training a central/global machine learning model in a distributed machine learning system. In the data sampling techniques, a subset of the data obtained at the local sites is intelligently selected for transfer to the central site for use in training the central machine learning model. In the model merging techniques, distributed local training occurs in each local site and copies of the local machine learning models are sent to the central site for aggregation of learning by merging of the models. As a result, in accordance with the examples presented herein, a central machine learning model can be trained based on various representations/transformations of data seen at the local machine learning models, including sampled selections of data-label pairs, intermediate representation of training errors, or synthetic data-label pairs generated by models trained at various local sites.
Abstract:
In one embodiment, a method includes: determining a plurality of candidate paths for a plurality of media streams, where each of the candidate paths is characterized by a first set of performance attributes and each of the plurality of media streams is characterized by a set of stream parameters; jointly determining a respective path from among the plurality of candidate paths that satisfies the set of stream parameters for each of the plurality of the media streams; and coordinating transmission of the plurality of media streams via the jointly determined respective path for each of the plurality of media streams. According to some implementations, the method is performed by a device with one or more processors and non-transitory memory.
Abstract:
Various implementations disclosed herein enable client devices to share a network resource in order to produce more evenly distributed perceptual playback quality levels within each subscription tier. Sharing of a network resource is facilitated by providing client devices with one or more QoE distribution indicator values, which enables client device participation in the allocation of the network resource. In some implementations, a client device method includes determining a local QoE level value, and then modifying a representation selection of media content data based on the local QoE and one or more QoE distribution indicator values. The local QoE level value characterizes perceptual playback quality of media content data received by the client device using the shared network resource. The QoE distribution indicator values characterize at least in part the relative QoE values associated with a plurality of client devices sharing the shared network resource with the client device.
Abstract:
Various implementations disclosed herein enable a more efficient allocation of one or more shared network resources (e.g., bandwidth, memory, processor time, etc.) amongst a number of client devices based on media content data complexity and client device resource constraints in order to better manage perceptual playback quality of adaptive streaming content. In some implementations, a method includes aligning sequences of one or more temporal segments; and, selecting segment representations for each temporal segment based on a combination of the sequence alignment and perceptual quality level values associated with available segment representations, such that a combination of resulting perceptual quality levels satisfies a joint quality criterion. Each sequence is associated with one of a number of client devices sharing a network resource. The one or more temporal segments of each sequence are used to provide segment representations of media content data to one of the client devices.
Abstract:
An example method is provided and includes determining a time shift comprising a difference in time between a packet count observation at a transmit element and a corresponding packet count observation at a receive element connected to the transmit element via a link; obtaining a first packet count from the transmit element and a second packet count from the receive element; and adjusting at least one of the first packet count and the second packet count to compensate for the time shift. The method further includes comparing the adjusted first and second packet counts to determine whether there is a discrepancy between the counts and if a discrepancy is detected between the counts, adjusting a rate at which the transmit element sends packets to the receive element.
Abstract:
Embodiments include technologies for creating a manifest for a conferencing event in a network, adding a name tag identifying the conferencing event to the manifest, receiving an interest packet including one or more parameters indicating a named flow being produced at a source node, adding content metadata of the named flow to the manifest, and sending the manifest to the source node. Further embodiments include adding, to the manifest, session-level metadata associated with a user of the source node. Embodiments include receiving a second interest packet with one or more second parameters identifying a user of a client node, where the second interest packet indicates a request to authorize the user of the client node to subscribe to the conferencing event. In further embodiments, session-level metadata associated with the user is added to the manifest if the user is authorized to subscribe to the conferencing event.
Abstract:
In one embodiment, an HTTP streaming session may be initiated at a client device in a network. The client device may have a buffer and may be configured to request and receive one or more data segments over HTTP from an HTTP server. A first data segment at a first data source rate may be requested and subsequently received. The first data segment may be stored in the buffer. A second data source rate may then be calculated based on a storage level in the buffer, and a second data segment at the second data source rate may be requested.
Abstract:
A method includes establishing communication channels between an on-board unit (OBU) of a vehicle and a plurality of nodes, tagging each of a plurality of data from the plurality of nodes with a priority level, storing the plurality of data in a priority queue according to respective priority levels, selecting a medium to present a first data of the plurality of data to a user, and presenting the first data to the user via the medium. In the method, the plurality of nodes includes a remote node and an in-vehicle device. Another method includes receiving a data from a remote node, generating a plurality of data streams from the data and transmitting the plurality of data streams across a plurality of wireless interfaces. Another method includes enhancing audio signals from a plurality of microphones and speakers. Yet another method includes various gesture based user interfaces coupled to the OBU.