Invented by Yi Zhu, Kiran Kumar Edara, Kaixiang Hu, Andrew David Price, Amazon Technologies Inc

The market for server-side rate control for adaptive bitrate streaming protocols has been steadily growing in recent years. With the increasing popularity of video streaming services and the demand for high-quality video content, the need for efficient and effective rate control solutions has become paramount. Adaptive bitrate streaming protocols, such as Dynamic Adaptive Streaming over HTTP (DASH) and HTTP Live Streaming (HLS), have become the industry standard for delivering video content over the internet. These protocols allow for seamless playback by adjusting the video quality in real-time based on the viewer’s network conditions. This ensures a smooth viewing experience without buffering or interruptions. Server-side rate control plays a crucial role in adaptive bitrate streaming by dynamically adjusting the video bitrate based on various factors such as network congestion, device capabilities, and viewer preferences. This allows for optimal video quality while minimizing bandwidth usage and reducing the risk of buffering. The market for server-side rate control solutions is driven by the increasing demand for high-quality video streaming across various devices and platforms. As more and more viewers consume video content on smartphones, tablets, smart TVs, and other devices, the need for adaptive bitrate streaming solutions that can deliver consistent and high-quality video experiences becomes crucial. Moreover, the rise of over-the-top (OTT) streaming services has further fueled the demand for server-side rate control solutions. OTT platforms, such as Netflix, Amazon Prime Video, and Disney+, rely heavily on adaptive bitrate streaming to deliver their vast libraries of content to millions of viewers worldwide. These platforms require robust rate control algorithms that can efficiently manage the streaming quality for a large number of concurrent viewers. The market for server-side rate control solutions is also driven by the increasing complexity of video codecs and formats. New video codecs, such as High Efficiency Video Coding (HEVC) and AV1, offer better compression efficiency and improved video quality. However, these codecs require sophisticated rate control algorithms to ensure optimal video quality and efficient bandwidth utilization. Several companies are actively involved in the development and commercialization of server-side rate control solutions. These companies offer a range of products and services that cater to the needs of video streaming platforms, broadcasters, and content delivery networks (CDNs). These solutions typically include advanced rate control algorithms, real-time monitoring and analytics, and integration with existing streaming infrastructure. The market for server-side rate control solutions is expected to witness significant growth in the coming years. The increasing demand for high-quality video streaming, the proliferation of OTT platforms, and the advancements in video codecs and formats are expected to drive the adoption of server-side rate control solutions across the industry. In conclusion, the market for server-side rate control for adaptive bitrate streaming protocols is experiencing steady growth due to the increasing demand for high-quality video streaming and the complexity of video codecs and formats. As the industry continues to evolve, server-side rate control solutions will play a crucial role in delivering optimal video quality and a seamless viewing experience to viewers worldwide.

The Amazon Technologies Inc invention works as follows

Network Hardware Devices organized in a Wireless Mesh Network (WMN), where one mesh network device consists of a first radio, and one or several additional radios connected to an application processor. The application processor receives, via the 1st radio, a request for streaming content data to the client consumption device. It then receives, via one or more additional Radios, portions of content data from different devices, at different retrieval speeds. The application processor determines the desired streaming bitrate, calculates the average retrieval rate of the first and second segments, and decides the end time for transmission to the client consumption device. The application processor transmits the first part via the first radio such that the at least one bit of the portion is transmitted by the end of transmission.

Background for Server-side Rate Control for Adaptive Bitrate Streaming Protocols

A growing number of people are enjoying digital media such as movies, music, images, electronic book, etc. Users use a variety of electronic devices to consume media. These electronic devices, also known as user equipment or user devices, include electronic book readers (also called PDAs), cellular phones, portable media players (PMPs), tablet computers, netbooks and laptops. These electronic devices communicate wirelessly with a communication infrastructure to allow the consumption of digital media items. These electronic devices have one or more antennas to allow them to communicate wirelessly with other devices.

BRIEF DESCRIPTION DES DRAWINGS

The following detailed description and accompanying drawings are intended to help you understand the inventions. They should not be interpreted as limiting the invention to specific embodiments.

FIG. “FIG. 1 is a diagram of a wireless mesh (WMN) network of network hardware devices for distributing content to client devices, in an environment with limited connectivity to broadband Internet infrastructure.

FIG. “FIG.

FIG. “FIG.

FIG. “FIG.

FIG. According to one embodiment, FIG. 5 shows a WMN that has twelve network hardware devices with four radios each. These form a backbone network of wireless peer-to-peer connections (P2P).

FIG. “FIG.

FIG. “FIG.

FIG. “FIG.

FIG. “FIG.

The description of a wireless mesh network (WMN), which contains multiple mesh networks devices organized in a topology mesh, is given. The mesh network device in the WMN collaborates in distributing content files to client consumption units in an environment with limited connectivity to broadband Internet Infrastructure. In developing nations there may be a lack of or slow roll-out of broadband internet infrastructure. The mesh networks described here can be used as a temporary solution until broadband internet infrastructure is widely available in developing nations.

Disclosed herein are embodiments relating to server-side rates control for adaptive bitrate stream protocols. As described in this document, the server-side control can be used to control data traffic between client consumption devices and mesh network devices. A mesh network device consists of a first and/or additional radios that are coupled to an application processing unit. The application processor receives, via the 1st radio, a request for content data streaming to a client consumption unit. It then receives, via one or more additional Radios, portions of content data from different devices, at varying retrieval rates. The application processor, for example, receives the first portions of content data via the second radio from a device and the second portions via the third radio from a device. The application processor determines the desired streaming bitrate, calculates the retrieval speed (e.g. average content retrieval rate) and the end time for transmission of the first portion. The application processor transmits the first part via the radio so that at least one bit of the portion is transmitted by the time the transmission is complete. Network hardware devices may also be referred to as mesh routers or mesh network devices. The network backbone is formed by multiple peer to peer (P2P), wireless connections. Wireless connections from node to client (N2C), or between multiple network devices and one or more consumer devices, are used to wirelessly connect the devices. Cellular connections are used to wirelessly connect multiple network devices to a Mesh Network Control Service (MNCS). Content files (or, more generally, a content item, object, or file) can be in any format of digital content. For example, electronic text (e.g. eBooks, electronic magazine, digital newspaper, etc.). ), digital audio (e.g., music, audible books, etc. ), digital video (e.g., movies, television, short clips, etc. ), images (e.g., art, photographs, etc. The client consumption devices may include any type of content rendering device such as electronic book readers, portable digital assistants, mobile phones (e.g. art or photographs), tablet computers, laptop computers with portable media players and cameras. Client consumption devices can include any kind of content rendering device, such as electronic books readers, portable digital devices, mobile phones and tablets, laptops, portable media players or tablet computers, cameras, videos cameras, netbooks, notebooks desktop computers gaming consoles DVD players media centers etc.

The mesh network devices can be deployed in environments with limited connectivity to broadband Internet infrastructure. In some embodiments, the mesh architecture described herein does not include “gateway” nodes. Nodes capable of forwarding mesh broadband traffic to the Internet are included in some embodiments. A limited number of POP nodes may have Internet access, but most mesh network devices are capable of forwarding mesh traffic to other mesh network devices to deliver content to clients that otherwise would not have broadband Internet connections. Alternatively, POP nodes can be coupled with storage devices to store content available for the WMN. The WMN can be self-contained, in that the content is stored, transported, and consumed by the nodes of the mesh network. In some embodiments of the mesh network architecture, a large number mesh nodes are used, called Meshbox Nodes. Hardware-wise, the Meshbox is similar to an enterprise router. It also has the capability of P2P connections. This forms the network backbone for the WMN. The Meshbox nodes are able to provide many of the capabilities of a CDN, but on a more localized basis. The WMN is able to be deployed in areas where broadband Internet access is not available. The WMN is scalable to support a geographical area based on number of mesh networks devices and distances that mesh network devices must travel in order to communicate successfully over WLAN channels.

Although the various embodiments described herein are directed at content delivery (such as the Amazon Instant Video service), the WMNs and corresponding mesh networks devices can be used to deliver high bandwidth content in applications where low latency or predictable access patterns are not critical. The embodiments described in this document are compatible with current content delivery technologies and may leverage architectural options, such as CDN service providers like Amazon AWS CloudFront. Amazon CloudFront CDN, a global CDN that integrates with Amazon Web Services products, distributes content with high data transfer rates and low latency to users. The embodiments described in this document can be used as an extension of the global CDN but for environments with limited broadband internet infrastructure. These embodiments may offer users in these environments a content delivery service that is comparable to what they would get on a broadband Internet connection. The embodiments of the present invention may be used to optimize the deployment for certain traffic types, such as streaming video. “Streaming video” is a growing part of broadband traffic, and it’s taxing the existing infrastructure.

FIGS. The 1-4 are directed at network hardware devices organized in a mesh wireless network for the distribution of content to client consumption devices when there is limited connectivity to broadband infrastructure. These network hardware devices may implement the embodiments described in this document. FIGS. 5-7B are aimed at routing traffic through a backbone network of the WMN mesh network devices. FIGS. “Figures 8-9 show devices with multi-radio multi-channel (MRMC), mesh networks that can implement the various embodiments described in this document.

FIG. According to one embodiment, FIG. 1 shows a diagram of network devices 102-110 organized in a Wireless Mesh Network (WMN) for content distribution in an environment with limited connectivity to broadband Internet infrastructure. The WMN includes multiple network devices 102-110 which connect together in order to transfer digital contents through the WMN to be delivered to one of more client consumption devices that are connected to the WMN. In the embodiment depicted, the WMN 100 comprises a miniature point of presence (mini-POP), also known as mini-POP Device. It has at least one wired connection with an attached storage device 103 and a wireless point-to point connection 105 with a CDN device 107 of an Internet Service Provider. The CDN device may be a point-of-presence (mini-POP) device (also known as mini POP device), a server edge, a content server or another device from the CDN. In operation, the mini-POP 102 is similar to CDN POP devices. The mini-POP 102 device is called a mini to distinguish it from a CDN POP Device, as the mini POP 102 device only has one ingress point into the WMN 100, whereas the CDN POP Device may have many.

The point to point wireless connection 105 can be established using a point to point wireless link 115, between the mini POP device 102 and CDN device. The point-to-point connection 105 can also be established via a microwave directional link between the CDN device and the mini-POP. In some embodiments, the WMN’s 100 content files are accessed by the mini-POP 102. The mini-POP 102 could be the only WMN node with access to an attached storage device or a communications channel for retrieving content files outside the WMN. In other embodiments multiple mini-POPs may be deployed within the WMN 100. However, the number of mini POPs should be smaller than the total number of network devices in the WMN 100. In some embodiments, a wireless point-to-point connection may be used. To exchange data, for example, you can use a microwave channel. There are other long-distance communication channels that can be used. These include fiber-optic links, satellite links, cellular connections, etc. The network hardware devices in the WMN 100 do not necessarily have direct access the mini POP device 102 but they can use one of more intermediary nodes to obtain content. Intervening nodes can also cache content so that other nodes can access it. The network hardware devices can also determine the shortest route between a node that is requesting content and another node that stores it.

The CDN device 107 could be located in a datacenter 119, and connected to the Internet 117. The CDN device may be just one of the many devices that make up the global CDN, and it may use Amazon CloudFront. The CDN device and datacenter 119 can be located in the same location as the equipment for the point-topoint wireless link. The point-topoint wireless connection 105 is a broadband connection that can be used for the WMN 100. In some cases the mini-POP device has no Internet connection through the point-topoint wireless connection 105, and content is only stored in the attached storage device for a self contained WMN 100.

Click here to view the patent on Google Patents.