Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Group-Enabled, Temporally Scalable, Rate-Control

IP.com Disclosure Number: IPCOM000193447D
Original Publication Date: 2010-Feb-24
Included in the Prior Art Database: 2010-Feb-24
Document File: 2 page(s) / 193K

Publishing Venue

Motorola

Related People

Tyrone Bekiares: INVENTOR [+4]

Abstract

The authors proposed a simple, temporally-scalable, group-enabled, rate-control algorithm which will dynamically match the source's encoded bit rate to that of the receiver in the group who is subject to the lowest throughput at any given instant in time. As a result, all receivers in a group will receive and decode the same video stream, where the inherent temporal quality (i.e. frame rate) of that stream is inherently linked to the lowest throughput available to any given receiver in the group. To achieve this, the authors begin with a simple "frame pull" rate-control algorithm known in the art. For this particular rate-control algorithm, the resolution and quantizer used by the encoder are fixed in value. The encoder then generates encoded frames at a rate dictated by the receiver: when the receiver has successfully received an entire frame from the encoder, it "pulls" the next frame from the encoder. In this way, the throughput to the receiver is never exceeded. To enable this rate-control algorithm in a group environment, the authors propose that all of the receivers in the group forward their respective "frame pull" directives to a centralized entity, herein referred to as the Media Reflector. When the Media Reflector has determined that all receivers in the group have successfully received the previous frame, it then forwards a single "frame pull" directive back to the source encoder.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Problem Background

Matching the encoded bit rate of compressed video to the available throughput between an encoder and decoder is one of the primary issues associated with wireless video streaming. This problem is exasperated when multiple receivers are attempting to decode the same encoded bit stream, as the path to each receiver may be subject to vastly differing throughputs. This is particularly true if the receivers are connected via heterogeneous networks.

Proposed Solution

The authors proposed a simple, temporally-scalable, group-enabled, rate-control algorithm which will dynamically match the source’s encoded bit rate to that of the receiver in the group who is subject to the lowest throughput at any given instant in time. As a result, all receivers in a group will receive and decode the same video stream, where the inherent temporal quality (i.e. frame rate) of that stream is inherently linked to the lowest throughput available to any given receiver in the group.

To achieve this, the authors begin with a simple “frame pull” rate-control algorithm known in the art. For this particular rate-control algorithm, the resolution and quantizer used by the encoder are fixed in value. The encoder then generates encoded frames at a rate dictated by the receiver: when the receiver has successfully received an entire frame from the encoder, it “pulls” the next frame from the encoder. In this way, the throughput to the receiver is never exceeded.

To enable this rate-control algorithm in a group environment, the authors propose that all of the receivers in the group forward their respective “frame pull” directives to...