Adaptive Video Streaming Bit Rate Control and Version Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dynamic adaptive video streaming technologies face challenges in efficiently allocating limited computation and bit rate resources for encoding multiple video versions, leading to high complexity and poor resource utilization, especially in real-time encoding scenarios where pre-encoding is not feasible.
Innovation Solution
A method and system for bit rate control and version selection that uses a dynamic adaptive streaming media encoding technology, employing an encoding complexity-bit rate-distortion model to optimize video version encoding, considering server constraints and user network conditions, to maximize video quality while efficiently using resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple video versions are encoded to adapt to different user needs and network conditions, then user satisfaction is improved, but server computation resources are excessively consumed
Solution Approach 1:
The patent changes the parameter of video encoding by pre-determining the optimal number of versions and their specific bit rates based on historical data and prediction algorithms, rather than encoding all possible versions. This parameter optimization reduces server computation resources while maintaining video version adaptability for different user needs and network conditions.
Solution Approach 2:
The patent applies preliminary action by predicting future video encoding demands and pre-determining the optimal version configurations before actual encoding occurs. This allows the system to prepare encoding parameters in advance, reducing real-time computation resources while maintaining the ability to adapt to different user requirements.
2Ease of operation
If pre-encoding is performed to obtain all video versions, then version selection is simplified, but encoding time and computation resources are excessively occupied
Solution Approach 1:
The patent optimizes encoding parameters by determining the optimal number of video versions and their bit rates in advance based on prediction algorithms, rather than performing exhaustive pre-encoding. This parameter optimization maintains ease of version selection while significantly reducing the time and computation resources required for encoding.
Solution Approach 2:
The patent applies partial action by encoding only the necessary number of video versions that are predicted to be needed, rather than encoding all possible versions. This selective approach maintains sufficient version selection capability while reducing encoding time and computation resource consumption.
3Adaptability or versatility
If exhaustive version encoding is performed, then all user needs are covered, but bit rate resource allocation becomes inefficient
Solution Approach 1:
The patent optimizes bit rate resource allocation by determining the optimal number of video versions and their specific bit rates in advance based on prediction algorithms and historical data. This parameter optimization ensures comprehensive coverage of user needs while efficiently allocating bit rate resources, avoiding waste from encoding unnecessary versions.
Solution Approach 2:
The patent applies local quality by allocating different bit rates to different video versions based on predicted demand and user requirements. This allows the system to concentrate bit rate resources on the most needed versions rather than uniformly distributing resources across all possible versions, improving overall efficiency while maintaining adaptability.
Data Source
AI summary
The disclosure provides a method and system for encoding bit rate control and version selection for a dynamic adaptive video streaming media. The method adopts a dynamic adaptive streaming media encoding technology to encode each original video into a plurality of versions with different bit rates at a server and determines video version subsets to be encoded by the original videos and specific encoding parameters of each video version by taking an encoding complexity-bit rate-distortion model for different original video contents, constraints on an encoding bit rate and a computing resource of the video server, network connection conditions of different users and a video-on-demand probability distribution into consideration, and finally, the video server outputs an optimal video version set through encoding, so as to maximize the overall quality of videos watched by users.