Pipeline Data Exchange via Feedback Mechanism

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing pipeline-based multimedia frameworks face challenges in obtaining internal information during data processing, leading to low reusability of data streams and requiring pipeline restarts when parameter adjustments are needed.

Innovation Solution

A data exchanging method involving a generic element and a docking element within a pipeline, where the generic element processes raw data to generate analysis data, the docking element processes this analysis data to generate feedback data, and the generic element adjusts subsequent data processing based on this feedback, allowing for dynamic data exchange and online parameter adjustments without restarting the pipeline.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a pipeline-based multimedia framework is used with interchangeable elements, then data exchange is easy to achieve, but internal information cannot be obtained during execution and data stream reusability is low

Engineering Contradiction:

Improvedata exchange easeVSAvoiddata stream reusability

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The pipeline element is segmented into multiple processing stages (first processing unit, second processing unit, third processing unit) that can independently process data streams. This segmentation allows different parts of the pipeline to handle different aspects of data processing, enabling the same data stream to be reused across multiple processing paths simultaneously.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A feedback data structure is introduced as an intermediary mechanism that carries internal information between processing units. This feedback mechanism enables downstream elements to obtain internal information from upstream elements without requiring pipeline restart, thus improving data stream reusability while maintaining ease of operation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If the pipeline is executed, then data processing is continuous, but internal information cannot be obtained during the process

Engineering Contradiction:

Improvedata processing continuityVSAvoidinternal information accessibility

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

A feedback mechanism is implemented where processing results and internal information are fed back from downstream elements to upstream elements during pipeline execution. The feedback data structure includes fields for storing processing results, error information, and control parameters, enabling continuous information exchange without interrupting the data processing flow.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If an associated parameter needs to be revised during pipeline execution, then parameter flexibility is improved, but the pipeline must be restarted

Engineering Contradiction:

Improveparameter adjustabilityVSAvoidpipeline restart time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The pipeline parameters are made dynamic through the feedback mechanism, allowing parameters to be adjusted during runtime without restarting the pipeline. The feedback data structure carries control parameters that can modify processing behavior on-the-fly, enabling the pipeline to adapt to changing requirements while maintaining continuous operation.

Inventive Principle:

Principle #15Dynamics

4Ease of manufacture

If elements are interchangeable and quickly combined, then development convenience is improved, but data stream reusability remains low

Engineering Contradiction:

Improvedevelopment convenienceVSAvoiddata stream reusability

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The pipeline elements are designed with universal interfaces and standardized data structures that enable them to serve multiple functions. The feedback data structure is a universal mechanism that can be used across different element types, allowing the same data stream to be reused across multiple processing paths while maintaining development convenience through standardized interfaces.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250156191A1Data exchanging method of pipeline elements and non-transitory computer readable medium

Publication Date: 2025.05.15 ADLINK TECH INC

AI summary

A data exchanging method of pipeline elements is adapted to process a pipeline of a data stream, and includes: processing, by a generic element in the pipeline, first raw data in the data stream to generate first analysis data; receiving and processing, by a first docking element, the first analysis data to generate the; and receiving, by the generic element, data associated with first feedback data and processing, by the generic element, second raw data in the data stream according to the data associated with first feedback data.