Video Transmission Frame Selection for Latency Stabilization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional video transmission technologies face challenges in stabilizing latency and jitter due to dynamic source-channel mismatches, particularly in real-time wireless communication scenarios like unmanned aerial vehicle applications, where sudden changes in channel bandwidth or source encoding cause fluctuations in transmission time.

Innovation Solution

The method involves dividing an image/video sequence into subsequences, encoding them separately, and selecting frames for transmission based on each subsequence's bitstream size and current channel bandwidth, with error frames reconstructed using linear interpolation, ensuring stable and fault-tolerant transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional rate control algorithms are used to stabilize average bit rate, then overall average latency is controlled, but per-frame latency and jitter cannot be resolved due to dynamic source-channel mismatch

Engineering Contradiction:
Improvetransmission stabilityVSAvoidper-frame latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The video stream is segmented into multiple channels with different code rates. The transmitting terminal divides the encoded video data into multiple channels, each with different coding rates, allowing the receiving terminal to select appropriate channels based on current channel conditions, thereby resolving per-frame latency and jitter issues that conventional average-based rate control cannot address

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically adapts to changing channel conditions by selecting from multiple pre-encoded channels with different code rates. Instead of using a fixed or slowly-adapting average bit rate, the system can immediately switch between channels to match current bandwidth availability, resolving the dynamic source-channel mismatch problem

Inventive Principle:
Principle #15Dynamics

2Quantity of substance

If the channel bandwidth is reduced by one half, then transmission capacity decreases, but the transmission latency is doubled accordingly

Engineering Contradiction:
Improvechannel bandwidthVSAvoidtransmission latency
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The video data is pre-encoded into multiple channels with different code rates before transmission. When channel bandwidth changes, the receiving terminal can immediately select the appropriate pre-encoded channel without waiting for re-encoding, thus avoiding latency doubling when bandwidth is reduced

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If a camera moves abruptly or an object in camera view has large motion, then encoded bitstream size is suddenly doubled, but transmission latency is doubled accordingly

Engineering Contradiction:
Improvemotion adaptabilityVSAvoidtransmission latency
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The video stream is divided into multiple channels with different code rates to handle varying motion complexity. When abrupt motion occurs causing bitstream size to double, the system can switch to a higher code rate channel that was pre-encoded, avoiding the latency penalty of real-time rate adjustment

Inventive Principle:
Principle #1Segmentation

4Reliability

If multiple channels of coded data are prepared for the same data, then fault tolerance is improved, but device complexity increases

Engineering Contradiction:
Improvefault toleranceVSAvoidencoding complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The video data is segmented into multiple channels with different code rates using separate encoding processes. This segmentation provides fault tolerance by allowing selection of appropriate channels based on channel conditions, while the modular structure manages encoding complexity through independent channel encoders

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3681156B1Method and device for transmitting wireless data
Publication Date: 2021.07.14 SZ DJI TECH CO LTD
  • EP3681156B1 patent drawingFigure 1~2
  • EP3681156B1 patent drawingFigure 3~4
  • EP3681156B1 patent drawing

AI summary

A method for processing video data is provided. The method includes down-sampling temporally an image sequence to form a plurality of subsequences; encoding the plurality of subsequences separately to form a plurality of encoded subsequences; and selecting frames for transmission according to a size of each frame in the plurality of encoded subsequences and a channel bandwidth. An unmanned aerial vehicle and computer-readable storage medium are also provided.