At&t Intellectual Property I, L.p.

At&t Mobility Ii LLC

University of Illinois

Georgia Tech Research Corporation

University College Cork

The Governors of the University of Calgary

University of Southern California

Old Dominion University Research Foundation

Emir Halepovic

Vijay Gopalakrishnan

Jeffrey Erman

Rittwik Jana

Rakesh Sinha

Subhabrata Bapi Sen

Xuan Kelvin Zou

Simon D Byers

Christopher Volinsky

Eric Petajan

Carlos Eduardo De Andrade

Lien K Tran

Matteo Varvello

David John Poole

Hulya Seferoglu

Muhammad Usama Chaudhry

Shanyu Zhou

Balajee Vamanan

Sarat Puthenpura

Cheuk Yiu Ip

Balagangadhar Bathula

Cormac John Sreenan

Marco Platania

Kyung-Wook Hwang

Harold Orr McCaskey

Ellen Witte Zegura

Mostafa H Ammar

Ahmed Zahran

Niklas Carlsson

Shibin Mathew

Vengatanathan Krishnamoorthi

Tarun Mangla

Darijo Raca

Theodore Karagioules

Mea Wang

Kadangode K Ramakrishnan

Jennifer Yates

Robert C Streijl

Shuai Hao

Gowrishankar Natarajan

Pankaj Sharma

Ramesh Govindan

Shichang Xu

Atanu Basak

Robert Zegheru

Constance Goshgarian

Yaguang Li

Albert Chan

Cyriac James

Edward Allen Lambert

Ibrahim Ben Mustafa

Zahaib Akhtar

Yang Liu

Rachit Sharma

Tamer Nadeem

Aspects of the subject disclosure may include, for example, adjusting bandwidth allocated for a first group of communication devices communicatively coupled to a first communication network from a bandwidth limited allocation to a no bandwidth limited allocation, and measuring a first group of metrics associated with the first group of communication devices associated with the first communication network. Further embodiments include computing a second group of metrics associated with a second group of communication devices communicatively coupled to a second communication network according to the first group of metrics, the second group of communication devices have the no bandwidth limited allocation, and allocating a first group of network resources for the second group of communication devices according to the second group of metrics. Other embodiments are disclosed.

Methods, devices, and systems for estimating video data usage without bandwidth limits

A method includes acquiring a playlist identifying a plurality of video files to be played back during a video streaming session on an endpoint device, obtaining, from an operator of a communications network over which the video files are to be downloaded, a maximum throughput for the video streaming session, selecting a subsequent video file that is scheduled in the playlist for playback after a playback of a currently playing video file, wherein the subsequent video file includes a plurality of chunks, selecting, based on the maximum throughput and for each of the chunks, a variant, such that a plurality of variants is selected in which each variant corresponds to one chunk of the plurality of chunks, wherein the variants are selected to maximize a utility function that is constrained by the maximum throughput, and downloading the plurality of variants to a local buffer of the endpoint device.

Optimization of bitrate adaptation and prefetching for short-form video

A method includes acquiring a playlist identifying video files and an order in which the video files are to be played back during a video streaming session on an endpoint device, obtaining, from an operator of a communications network over which the video files are to be downloaded, a maximum value of a streaming parameter for the video streaming session, selecting a subsequent video file that is scheduled for playback after a playback of a currently playing video file, wherein the subsequent video file includes a plurality of chunks, selecting, for each chunk, a variant, such that a plurality of variants is selected in which each variant corresponds to one chunk of the plurality of chunks, wherein the selecting the variant is based on both the maximum value and a measured value for the streaming parameter, and downloading the plurality of variants to a local buffer of the endpoint device.

Bitrate adaptation and prefetching for short-form video

Aspects of the subject disclosure may include, for example, obtaining predicted available bandwidths for an end user device, monitoring buffer occupancy of a buffer of the end user device, determining bit rates for portions of media content according to the predicted available bandwidths and according to the buffer occupancy, and adjusting bit rates for portions of media content according to the predicted available bandwidths and according to the buffer occupancy during streaming of the media content to the end user device over a wireless network. Other embodiments are disclosed.

Method and system for managing service quality according to network status predictions

Devices, computer-readable media, and methods for making a cache admission decision regarding a video chunk are described. For instance, a processing system including at least one processor may obtain a request for a first chunk of a first video, determine that the first chunk is not stored in a cache, and apply, in response to the determining that the first chunk is not stored in the cache, a classifier to predict whether the first chunk will be re-requested within a time horizon, where the classifier is trained in accordance with a set of features associated with a plurality of chunks of a plurality of videos. When it is predicted via the classifier that the first chunk will be re-requested within the time horizon, the processing system may store the first chunk in the cache.

Adaptive bitrate video cache

Network capacity planning based on application performance can include detecting a data session occurring on a network, identifying an application being used for the data session, where the application can include a video application, determining if a performance model for the video application exists, the performance model describing performance metrics and quality of service events associated with the video application, determining, based on the performance model, a capacity planning trigger for the video application, where the capacity planning trigger can include increasing network capacity based on the needs and a quality of service associated with the video application during the data session, and generating a command that, when executed by a network entity, causes the network entity to implement the capacity planning trigger on the network.

Network capacity planning based on application performance

Aspects of the subject disclosure may include, for example, obtaining, from a first source of information, a first bandwidth prediction, wherein the first bandwidth prediction is based upon historical bandwidth data that had been provided by a plurality of devices; obtaining, from a second source of information, a second bandwidth prediction, wherein the second bandwidth prediction is based upon network measurements, and wherein the network measurements are other than the historical bandwidth data that had been provided by the plurality of devices; selecting as a source of a future bandwidth prediction one of the first source of information and the second source of information, wherein the selecting is based upon a comparison of each of the first bandwidth prediction and the second bandwidth prediction to an actually obtained bandwidth of the device. Other embodiments are disclosed.

Apparatus and method for selecting a bandwidth prediction source

A method includes receiving, at a device from a content source, a portion of a particular chunk of media content a particular quality level during adaptive bit rate media streaming of the media content. The method includes receiving, at the device, network data related to network conditions associated with the device in response to receipt of the portion. The method also includes, in response to a determination to download the particular chunk at a first quality level different than the particular quality level, the determination based on the particular quality level and the network data: sending a request from the device for the particular chunk at the first quality level to the content source, and replacing the portion with the particular chunk at the first quality level.

Adaptive bit rate media streaming based on network conditions received via a network monitor

Concepts and technologies directed to agile transport for background traffic in cellular networks are disclosed herein. In various aspects, a system can include a processor and memory storing instructions that, upon execution, cause performance of operations. The operations can include determining a capacity of a communication path that communicatively couples a user equipment to a radio access network cell site. The operations can include identifying, from the radio access network cell site, a queue that is constructed for the user equipment. The operations can include assembling a plurality of probe burst packet sets from a background traffic flow. The operations can include probing the communication path for spare capacity using the plurality of probe burst packet sets and delivering the background traffic flow to the user equipment using the spare capacity while the communication path is not busy.

Agile transport for background traffic in cellular networks

Aspects of the subject disclosure may include, for example, setting a streaming rate cap for a client device; obtaining a bitrate ladder associated with the client device, wherein the bitrate ladder comprises a listing of a plurality of bitrates that the client device can request in connection with receiving streaming data; determining a boost rate for the client device, wherein the boost rate is greater than the streaming rate cap such that the boost rate enables the client device to receive the streaming data faster than the client device would otherwise receive the streaming data under the streaming rate cap, and wherein the boost rate is based upon the bitrate ladder; and enabling delivery to the client device of the streaming data up to the boost rate; wherein a network device is part of a network through which the streaming data is delivered to the client device. Other embodiments are disclosed.

Cap-based client-network interaction for improved streaming experience

Aspects of the subject disclosure may include, for example, a method in which a processing system identifies a plurality of performance indicators comprising device performance indicators for a plurality of communication devices on a cellular network and network performance indicators for the cellular network. The method also includes obtaining historical data regarding the plurality of performance indicators for each of a series of time points during a past time period; the historical data for each of the plurality of performance indicators form an array of values for that performance indicator. The method further includes generating from each array a set of inputs to an algorithm for predicting a throughput of the cellular network during a future time period; the set of inputs comprises quantiles of the array, and the algorithm comprises a machine learning algorithm. Other embodiments are disclosed.

System and method for throughput prediction for cellular networks

Aspects of the subject disclosure may include, for example, a method in which a processing system obtains encoded video content, and transmits the encoded video content over a network for presentation at a communication device by a video player executing on the communication device. The encoded video content is decoded in a decoding process by a decoder of an operating system of the communication device to generate decoded video content; the decoded video content is analyzed by a complexity analyzer of the operating system to generate a video complexity report. The method also includes receiving the video complexity report over the network from the communication device, and adjusting a data rate for the encoded video content, based at least in part on the video complexity report, to generate adjusted video content for transmission to the communication device. Other embodiments are disclosed.

Video complexity detection for network traffic management

Aspects of the subject disclosure may include, for example, selecting a first video service model according to a server hostname determined according to transport layer security (TLS) transaction data associated with a video session transmitted over a network, calculating a plurality of data delivery statistics from the TLS transaction data according to the first video service model, where the plurality of data delivery statistics includes session-level statistics over the video session, transaction-level statistics over each transaction, and temporal feature statistics over intervals of the video session, determining a quality of experience (QoE) metric for the video session from the plurality of data delivery statistics according to a course-grained data QoE model, and adjusting a first network element of the network responsive to the determining the QoE metric for the video session. Other embodiments are disclosed.

Method and apparatus for estimating quality of experience from network data

A processing system may determine, during a downloading of a first video chunk associated with a first time block of a video, a risk of stall condition, the first video chunk encoded at a first bitrate of a plurality of available bitrates, determine that a number of a received first plurality of frames of the first video chunk exceeds a threshold, transmit a request for a next video chunk associated with a next time block of the video and an indication to a source device to stop transmitting additional frames of the first video chunk, receive a second plurality of frames of the first video chunk from the source device that were transmitted prior to the source device receiving the indication, receive frames of the next video chunk, and play the first and second plurality of frames of the first video chunk, and the frames of the next video chunk.

Adaptive bitrate streaming stall mitigation

Growing community of inventors

Somerset, NJ, United States of America

Emir Halepovic