Caching Entity for Video Segments Using Bit Rate Histograms

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing caching systems for encoded video segments struggle to efficiently identify the subset of segments to cache based on popularity values and estimated fluctuations in available bit rates, leading to suboptimal video delivery quality and potential congestion.

Innovation Solution

A caching entity that generates a histogram of available bit rates and combines popularity values with histogram frequencies to rank encoded video segments, using a decision unit to identify the subset to cache, while a monitoring unit intercepts communication to derive congestion information and fine-tune bit rate estimates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the caching entity caches all encoded video segments, then the video delivery quality improves, but the network congestion increases and system complexity increases

Engineering Contradiction:
Improvevideo delivery qualityVSAvoidnetwork congestion
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent applies local quality by differentiating the caching strategy for different video segments based on their individual popularity values. Instead of uniformly caching all segments or none, the system selectively caches only those segments with popularity values above a threshold, optimizing cache resource allocation to deliver high quality for popular content while avoiding unnecessary caching of less popular segments that would contribute to congestion.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the parameter of cache selection by introducing a dynamic threshold mechanism. The threshold is adjusted based on cache capacity and network conditions, allowing the system to adaptively control which segments are cached. This parameter change enables the system to balance between video delivery quality and network congestion by modifying the caching criterion rather than using a fixed all-or-nothing approach.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the caching entity uses complex algorithms to identify segments to cache, then the video delivery quality improves, but the device complexity increases

Engineering Contradiction:
Improvevideo delivery qualityVSAvoidcaching decision complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-calculating and storing popularity values for each video segment before caching decisions are made. The popularity values are computed based on historical access patterns and segment characteristics, allowing the decision unit to make rapid caching decisions by simply comparing current requests against pre-computed popularity data rather than performing complex real-time analysis.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by maintaining popularity values as a simplified representation or copy of the actual segment usage patterns. Instead of analyzing complete request histories or complex access patterns in real-time, the system uses pre-computed popularity value copies that capture the essential caching decision information, significantly reducing computational complexity while maintaining effective caching performance.

Inventive Principle:
Principle #26Copying

3Measurement precision

If the caching entity monitors all network communications, then the bit rate estimation accuracy improves, but the device complexity and processing overhead increase

Engineering Contradiction:
Improvebit rate estimation accuracyVSAvoidmonitoring system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies taking out by extracting only the essential information needed for bit rate estimation from network communications. The monitoring unit selectively extracts bandwidth utilization data, segment request patterns, and quality level information from the communication stream, ignoring other irrelevant data. This extraction approach provides sufficient accuracy for caching decisions without requiring complete analysis of all communication details.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses partial action by implementing monitoring at strategic points in the network rather than comprehensively monitoring all communications end-to-end. The monitoring unit is positioned to capture key metrics at the caching entity interface, obtaining sufficient information for accurate bit rate estimation without the excessive overhead of full-network surveillance. This partial monitoring approach achieves the necessary measurement precision with reduced complexity.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2466870B1Caching entity
Publication Date: 2015.07.08 ALCATEL LUCENT SA
  • EP2466870B1 patent drawingFigure 1
  • EP2466870B1 patent drawingFigure 2
  • EP2466870B1 patent drawingFigure 3

AI summary

Caching entity adapted to cache encoded video segments for delivery to a plurality of client entities over a network using a request-response protocol such as the hypertext transport protocol, said video segments being part of a video object and each of said video segments being encoded according to a set of predefined respective quality levels, said caching entity including a popularity measurement unit adapted to maintain popularity values for respective ones of said encoded video segments and a decision unit adapted to identify in function of said popularity value a subset of encoded video segments to be cached by said caching entity, characterized in that said caching entity also includes a rate estimation unit adapted to obtain in function of time an estimated fluctuation in available bit rate over said network between said caching entity and said client entities and that said decision unit in addition to said popularity values also uses said estimated fluctuation to identify said subset of encoded video segments in such a way that said subset contains encoded video segments said user entities will most probably request.