Coded Multisource Media Format for Robust Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently streaming media data over networks, particularly in ensuring high availability and robustness while managing latency and bandwidth constraints.
Innovation Solution
The Coded Multisource Media Format (CMMF) is introduced, which encapsulates media data and employs forward error correction (FEC) encoding to generate repair objects. These objects are stored separately and can be retrieved from different physical server devices, enabling multi-source media streaming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If media data is streamed using traditional single-source methods, then delivery simplicity is maintained, but availability and robustness deteriorate when network issues occur
Solution Approach 1:
The media data is divided into multiple segments that can be delivered from different physical server devices. Each segment can be independently retrieved from different sources, allowing the system to fail over to alternative sources if one server fails or is unavailable. This segmentation enables high availability while maintaining manageable complexity through standardized retrieval operations.
Solution Approach 2:
A manifest file serves as an intermediary that provides network locations for media segments without requiring the client to maintain complex relationships with multiple servers. The manifest file abstracts the complexity of multi-source delivery, presenting a simplified interface where segments are retrieved from locations specified in the manifest, enabling reliable streaming without direct client-server complexity management.
2Reliability
If repair objects are stored on separate physical servers, then system robustness improves, but data retrieval complexity increases
Solution Approach 1:
The manifest file is prepared in advance with pre-configured network locations for both base media segments and repair objects. This preliminary configuration eliminates the need for runtime decisions about which servers to query, as the manifest already specifies the exact locations. Clients can directly retrieve repair objects from the predetermined locations without complex resolution logic, maintaining ease of operation while achieving high robustness.
3Reliability
If forward error correction encoding is applied, then media data robustness improves, but processing time and latency increase
Solution Approach 1:
The media data is segmented into smaller units that can be independently encoded with FEC. This allows parallel processing of multiple segments, reducing the total encoding time while maintaining the same level of robustness. The segmented approach enables progressive delivery where segments can be processed and transmitted in parallel, minimizing latency compared to encoding the entire media stream as a single large unit.
Data Source
AI summary
An example device for retrieving media data includes a memory configured to store media data; and a processing system including a decoder implemented in circuitry, the processing system being configured to: retrieve a manifest file indicating a plurality of network locations for at least a first segment of media data and a second segment of the media data, each network location among the plurality of network locations being hosted by a separate physical server device; retrieve at least a first portion of the first segment from a first network location of the plurality of network locations; retrieve at least a second portion of the second segment from a second network location of the plurality of network locations; and provide the at least first portion of the first segment and the at least second portion of the second segment to the decoder.