Predictive Media Streaming Control for Mobile Link Variability

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing media streaming technologies reactively adapt to variations in link quality, leading to performance limitations and increased likelihood of stalls and re-buffering, particularly in mobile networks with dynamic link conditions.

Innovation Solution

Leveraging link performance prediction (LPP) technology to proactively adjust media streaming parameters such as resolution, buffer level, and compression based on predicted network conditions, using machine learning and historical data to optimize media streaming performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If reactive adaptation to link quality variations is used, then device complexity is reduced, but reliability deteriorates due to increased stalls and re-buffering

Engineering Contradiction:

Improvemedia playback continuityVSAvoidstreaming system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by predicting future link quality conditions before they occur and proactively adjusting streaming parameters in advance. The link performance prediction module forecasts bandwidth variations, and the media player adjusts resolution and buffer levels beforehand to prevent stalls and re-buffering events, rather than reacting after problems occur.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If link performance prediction is implemented, then reliability improves by reducing stalls, but device complexity increases due to additional prediction and control mechanisms

Engineering Contradiction:

Improvemedia playback continuityVSAvoidstreaming system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system implements feedback mechanisms where the media player continuously monitors actual link quality and compares it with predicted values. This feedback loop allows the system to refine predictions and adjust streaming parameters dynamically, improving reliability while managing complexity through iterative optimization rather than overly complex predetermined controls.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The media streaming system performs self-service by autonomously predicting link quality and adjusting its own streaming parameters without external intervention. The link performance prediction module and media player work together to automatically optimize resolution, buffer levels, and pre-fetching strategies, reducing the need for complex external control mechanisms.

Inventive Principle:

Principle #25Self-service

3Productivity

If proactive parameter adjustment based on predictions is used, then productivity improves by preventing playback stalls, but loss of time increases due to prediction and processing overhead

Engineering Contradiction:

Improvemedia playback efficiencyVSAvoidprediction processing time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system applies partial action by adjusting only the most critical streaming parameters (resolution and buffer level) based on prediction confidence levels. When prediction accuracy is high, more aggressive parameter changes are made; when confidence is lower, more conservative adjustments are applied, balancing productivity improvement with minimal processing overhead.

Inventive Principle:

Principle #16Partial or excessive action

4Adaptability or versatility

If link performance prediction is used, then adaptability improves for dynamic network conditions, but measurement precision requirements increase

Engineering Contradiction:

Improvenetwork condition adaptationVSAvoidlink quality measurement accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system changes parameters by transforming link quality measurements into predicted future states using machine learning models. Instead of relying on precise real-time measurements alone, the system uses historical data patterns and statistical models to predict future bandwidth conditions, enabling adaptability to dynamic networks while reducing the immediate precision requirements of measurements.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12574429B2Link performance prediction and media streaming technologies

Publication Date: 2026.03.10 INTEL CORP

AI summary

In one embodiment, a computing device for receiving a media stream includes processing circuitry to receive a link performance prediction for a network link between the computing device and a network, which indicates a predicted performance of the network link during a future timeframe. Based on the link performance prediction, the processing circuitry identifies a performance objective for the media stream. The performance objective is associated with media stream content that will be received in the media stream over the network link for playback during the future timeframe. Based on the link performance prediction and the performance objective, the processing circuitry adjusts one or more media streaming parameters for the media stream content to be played during the future timeframe. The processing circuitry then receives the media stream content to be played during the future timeframe over the network link based on the media streaming parameter(s).