Hewlett-Packard Development Company, L.p.

Cisco Technology, Inc.

Wai-Tian Tan

John George Apostolopoulos

Xiaoqing Zhu

Bo Shen

Gene Cheung

Robert Edward Liston

Andrew J Patti

Susie J Wee

Herbert Michael Wildfeuer

Debargha Mukherjee

Michael Anthony Ramalho

Sumit Roy

Rob Liston

Michele Mae Covell

Mitchell D Trott

Michael Harville

Paul Edwin Jones

Pål-Erik Martinsen

Ashish Khisti

Jon Michael Snyder

John Ankcorn

Ahmed Badr

Chuohao Yeo

Kashyap Kodanda Ram Kambhatla

Herb Wildfeuer

Santosh Ghanshyam Pandey

Ramin Samadani

Takashi Sakamoto

Paul Anthony Polakos

Sujoy Basu

John Martin Graybeal

Mary G Baker

Antonio Ortega

Leonidas Kontothanassis

Mehdi Nikkhah

Tina Wong

Raj Kumar

Jakov Cakareski

Allen Miu

Mai-Hsuan Lu

John G Apostolopulos

Shi Su

Kaidong Wang

Ju Shen

John G c/o Hewlett-Packard Company Apostolopoulos

Susie J c/o Hewlett-Packard Company Wee

Sumit c/o Hewlett-Packard Company Roy

J G c/o Hewlett-Packard Company Apostolopoulos

Sumit c/o Hewlett-Packard Roy

Techniques for improved wireless reliability are provided. It is determined that a client device is at least one of an augmented reality (AR) or a virtual reality (VR) device. A default set of retry parameters and a second set of retry parameters are determined, where the second set of retry parameters result in increased wireless reliability, as compared to the default set of retry parameters. Data is transmitted to the client device using the second set of retry parameters.

Wireless reliability for devices in motion

A method for encoding a sequence of packets includes receiving the sequence of packets, generating a parity packet for a first group of packets within the sequence of packets, and transmitting the first group of packets and the parity packet. The parity packet is generated by performing an exclusive OR (XOR) operation over a plurality of packets in the first group of packets and at least one packet in a second group of packets. The second group is separated from the first group in the sequence by one or more packets.

XOR forward error correction for isolated and burst losses over a software-defined-wide area network

Techniques for improved wireless reliability are provided. It is determined that a client device is at least one of an augmented reality (AR) or a virtual reality (VR) device. A default set of retry parameters and a second set of retry parameters are determined, where the second set of retry parameters are more robust than the default set of retry parameters. Data is transmitted to the client device using the second set of retry parameters.

Multi-User Multiple Input, Multiple Output (MU-MIMO) data transmissions are provided with a forward-predictive precoding matrix to mitigate the effects of a change in a state of a communication channel. First and second soundings are performed, at first and second times, to a receive antenna over a channel and, responsive to each of the soundings, first and second Channel State Information (CSI) are received. Based on the first and second CSI, a change in a state of the channel over a time period between the first and second time is determined. Based on the change in the state of the channel, a forward-predictive channel state matrix and/or a forward-predictive precoding matrix are determined that reflect a state of the channel at a future time and that are consistent with the determined change in the state over the time period. The forward-predictive precoding matrix is applied to a data transmission.

Forward predictive precoded MIMO

Techniques for improved wireless reliability are provided. Data is transmitted to a client device using a first set of retry parameters. It is determined that the client device is an augmented reality (AR) or virtual reality (VR) device, and a second set of retry parameters is determined, where the second set of retry parameters are more robust than the first set of retry parameters. Data is transmitted to the client device using the second set of retry parameters.

A method is provided in a wireless access point in a wireless communications network. The method includes obtaining information characterizing a first wireless stream and the second wireless stream transmitted or received by the wireless access point. The information includes at least a wireless channel quality for each of the first wireless stream and the second wireless stream. The method further includes allocating transmission resources to the first wireless stream and the second wireless stream based on the obtained information. In response to a change in quality of the first wireless stream, the method further includes revising the allocation of transmission resources for the first wireless stream based on at least one of a target bit-rate and a target level of smoothness.

Adaptive resource allocation for media streams over wireless

A method is described and in one embodiment includes receiving at a first node in a communications network a Session Traversal Utilities for Network Address Translation ('STUN') message associated with a first flow, wherein the STUN message comprises a flow attribute including corresponding to the first flow; analyzing the flow attribute at the first node; setting policy corresponding to the first flow in the network based on the analyzing, wherein setting the policy includes using the flow attribute of the STUN message to configure a network path for the first flow in the communications network; and forwarding the STUN message to a next network node.

Techniques for informing communications networks of desired packet transport treatment

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.

Training distributed machine learning with selective data transfers

An access point (AP) is configured to transmit packets to a client device over a communication channel. The AP determines a motion indictor indicative of motion of the client device based on a sequence of channel state information measurements, and measures a signal-to-noise ratio (SNR). The AP selects a transmission parameter among candidate transmission parameters using a learning-based algorithm based on observation parameters including the motion indicator, the SNR, and a device identifier for the client device. The AP employs the transmission parameter to transmit packets to the client device, and measures a transmission performance associated with the transmission parameter based on the transmitted packets. The AP updates the learning-based algorithm based on the observation parameters and the transmission performance for a next pass through the selecting, the employing, and the measuring.

Learning-based wireless transmission parameter adaptation based on client activity detection

A method is described and in one embodiment includes receiving at a first node in a communications network a message associated with a first flow, wherein the message comprises a flow treatment attribute including metadata indicative of how the first flow should be treated in the network; analyzing the flow treatment attribute at the first node; setting policy for treatment of the flow in the network based on the analyzing; and forwarding the message to a next network node.

In various implementations, a method includes determining a sequence of source packets. In some implementations, the sequence of source packets satisfies a windowing condition. In various implementations, the method includes synthesizing a first set of one or more parity packets as a function of a first set of source packets in the sequence. In some implementations, the first set of source packets satisfies a first encoding pattern. In various implementations, the method includes synthesizing a second set of parity packets as a function of a second set of source packets in the sequence. In some implementations, the second set of source packets satisfies a second encoding pattern that is different from the first encoding pattern. In some implementations, the first and second encoding patterns characterize an encoding structure determined as a function of a channel characterization vector.

Malleable error control code structures suitable for adaptive error protection

An endpoint device receives a sequence of audio frames. The endpoint device determines for each audio frame a respective importance level among possible importance levels ranging from a low importance level to a high importance level based on content in the audio frame indicative of the respective importance level. The endpoint device associates each audio frame with the respective importance level, to produce different subsets of audio frames associated with respective ones of different importance levels. The endpoint device, for each subset of audio frames, applies forward error correction to a fraction of audio frames in the subset of audio frames, wherein the fraction increases as the importance level of the audio frames in the subset increases, and does not apply forward error correction to remaining audio frames in the subset.

Audio frame labeling to achieve unequal error protection for audio frames of unequal importance

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.

Method and device for improved multi-homed media transport

Presented herein are techniques for assignment of an identity to a group of captured images. A plurality of captured images that each include an image of at least one person are obtained. For each of the plurality of captured images, relational metrics indicating a relationship between the image of the person in a respective captured image and the images of the persons in each of the remaining plurality of captured images is calculated. Based on the relational metrics, a clustering process is performed to generate one or more clusters from the plurality of captured images. Each of the one or more clusters are associated with an identity of an identity database. The one or more clusters may each be associated with an existing identity of the identity database or an additional identity that is not yet present in the identity database.

Clustering-based person re-identification

Growing community of inventors

Sunnyvale, CA, United States of America

Wai-Tian Tan