Simulation-Based Video Encoding Ladder Evaluation for QoE
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video streaming technologies generate encoding ladders that result in sub-optimal streaming quality of experience (QoE) and storage footprint tradeoffs due to sequential enforcement of ladder constraints, missing opportunities to select encoded videos that satisfy multiple constraints simultaneously.
Innovation Solution
A simulation-based technique that formulates encoding ladder generation as a constrained optimization problem, concurrently accounting for multiple ladder constraints to identify and exploit opportunities for a single encoded video that improves the QoE/storage footprint tradeoff, using a parameterized objective function and simulation evaluation to optimize encoding ladders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If encoding ladders are generated using sequential heuristic enforcement of ladder constraints, then the encoding process is simple and computationally efficient, but the streaming QoE/storage footprint tradeoff is sub-optimal
Solution Approach 1:
The patent transforms the encoding ladder generation from a sequential heuristic process to a constrained optimization problem by changing the parameters and objective function. The objective function incorporates multiple ladder constraints (minimum bitrate, maximum bitrate, minimum quality difference, maximum quality difference) simultaneously, allowing the system to find optimal encoded video combinations that satisfy all constraints while optimizing the QoE/storage footprint tradeoff.
Solution Approach 2:
The patent replaces the mechanical sequential heuristic enforcement system with a simulation-based evaluation system. Instead of sequentially adding encoded videos based on simple heuristics, the system uses simulation to evaluate candidate encoding ladders against realistic network throughput traces and adaptive bitrate algorithms, providing a more accurate assessment of actual streaming performance.
2Reliability
If multiple ladder constraints are enforced simultaneously through optimization, then the streaming QoE/storage footprint tradeoff is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-generating a comprehensive set of candidate encoded videos with various resolutions and bitrates before the optimization process. This pre-computation allows the constrained optimization to work with a prepared candidate pool rather than generating videos on-the-fly, significantly reducing the computational complexity during the actual ladder generation phase.
Solution Approach 2:
The patent uses simulation copying to evaluate candidate encoding ladders. Instead of deploying and testing multiple encoding ladders in production environments, the system creates simulated copies of the streaming environment with realistic network throughput traces and executes adaptive bitrate algorithms in simulation, allowing rapid evaluation of multiple candidates without actual deployment overhead.
3Reliability
If a single encoded video is selected to satisfy multiple ladder constraints, then the storage footprint is reduced and QoE is improved, but the difficulty of finding such videos increases
Solution Approach 1:
The patent implements feedback through the simulation evaluation process. The simulation executes adaptive bitrate algorithms with realistic network throughput traces to measure actual streaming performance metrics (buffering events, quality switches, throughput utilization). This feedback information guides the optimization process to identify encoded video combinations that truly satisfy multiple constraints while delivering optimal QoE, making the detection of suitable videos systematic rather than guesswork.
Data Source
AI summary
In various embodiments, a simulation evaluation application generates a first streaming header based on rungs of a first candidate encoding ladder, where each rung specifies a resolution and a bitrate of a different encoded video. The simulation evaluation application executes an adaptive bitrate algorithm on the first streaming header based on a network throughput trace to determine a first value for a metric that is relevant to quality of experience. The simulation evaluation application generates a second streaming header based on a second candidate encoding ladder. The simulation evaluation application executes the adaptive bitrate algorithm on the second streaming header based on the network throughput trace to determine a second value for the metric. The simulation evaluation application compares the first value to the second value to determine that the first candidate encoding ladder instead of the second candidate encoding ladder should be used to stream the media title.