Edge Data Streaming Protocol for Video Analytics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional video compression techniques are ineffective in alleviating network bottlenecks in Multi-access Edge Computing (MEC) systems, particularly in data analytics pipelines where video stream data is processed for object recognition and other intensive computational tasks, leading to inefficiencies in data transmission and processing.

Innovation Solution

A data streaming protocol that identifies and transmits regions of interest from video frames instead of full frames, and generates and transmits inference data rather than raw video stream data, reducing the volume of data transmitted and processed, thereby conserving bandwidth and processing resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional video compression techniques are used to transmit video stream data, then data transmission efficiency is improved, but the data is optimized for viewing quality rather than analytics processing, resulting in ineffective bottleneck alleviation

Engineering Contradiction:

Improvedata transmission efficiencyVSAvoiddata optimization for analytics

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments video frames into regions of interest (ROI) and transmits only those segments to analytics servers. This segmentation approach reduces unnecessary data transmission while preserving the analytical value needed for object recognition and other analytics tasks, directly addressing the mismatch between conventional compression optimized for viewing and analytics requirements.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and transmits only the essential analytical information (inference data, object detections, metadata) from video streams rather than transmitting compressed video frames. This extraction principle eliminates the need for decompression at analytics servers and directly provides processed data suitable for analytics workloads, resolving the contradiction between transmission efficiency and analytics adaptability.

Inventive Principle:

Principle #2Taking out (Extraction)

2Measurement precision

If full video frames are transmitted to edge servers for analytics processing, then complete video data is available for analysis, but network bandwidth and processing resources are overwhelmed

Engineering Contradiction:

Improveanalytics processing accuracyVSAvoiddata transmission volume

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent performs preliminary processing of video data at the source or intermediate nodes to generate inference data, object detections, and metadata before transmission to edge servers. This preliminary action reduces the data volume requiring transmission while ensuring that analytics-ready information is available, maintaining processing accuracy without overwhelming network and computing resources.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transmits partial video information (only regions of interest, extracted inference data, and essential metadata) rather than complete video frames. This partial action approach provides sufficient data for analytics processing while dramatically reducing transmission volume, resolving the contradiction between analytics accuracy and data quantity.

Inventive Principle:

Principle #16Partial or excessive action

3Loss of energy

If video data is compressed for transmission, then bandwidth is conserved, but decompression processing is required at receiving servers, increasing processing complexity

Engineering Contradiction:

Improvebandwidth consumptionVSAvoiddecompression processing

Core Design Contradiction:

Loss of energyVSDevice complexity

Solution Approach 1:

Instead of compressing video data and then decompressing it at the receiving end, the patent inverts the approach by transmitting already-processed inference data and metadata that require no decompression. This inversion eliminates the decompression step entirely, conserving bandwidth through selective data transmission while reducing receiving server complexity.

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent replaces the mechanical decompression process with direct transmission of processed analytical information. By substituting the decompression mechanism with pre-processed data formats, the system reduces bandwidth consumption while eliminating the computational overhead of decompression at receiving servers.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11831698B2Data streaming protocols in edge computing

Publication Date: 2023.11.28 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Systems and methods are provided for reducing stream data according to a data streaming protocol under a multi-access edge computing. In particular, an IoT device, such as a video image sensing device, may capture stream data and generate inference data by applying a machine-learning model trained to infer data based on the captured stream data. The inference data represents the captured stream data in a reduced data size based on performing data analytics on the captured data. The IoT device formats the inference data according to the data streaming protocol. In contrast to video data compression, the data streaming protocol includes instructions for transmitting the reduced volume of inference data through a data analytics pipeline.