10-bit 4K Video Sample Packing for Bandwidth Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Processing 4K video signals with 10-bit samples poses challenges due to bandwidth limitations, as traditional systems struggle to efficiently handle the increased data requirements, particularly with 10 Gbit Ethernet ports, leading to inefficiencies in video processing and transmission.

Innovation Solution

A method for efficient 10-bit sample packing and unpacking is implemented, where 10-bit samples are stored in 16-bit words, with chroma samples split into lower and upper 5-bit portions and stored within the unused bits of luma samples, reducing bandwidth requirements and enabling efficient conversion between formats.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If 10-bit samples are used for 4K video signals to improve picture quality, then picture quality is improved, but bandwidth requirements increase

Engineering Contradiction:
Improvepicture qualityVSAvoidbandwidth requirements
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent embeds chroma sample data within the unused upper bits of luma sample containers. Each 16-bit word containing a 10-bit luma sample has 6 unused bits, which are utilized to store portions of chroma samples. This nesting approach allows simultaneous transmission of luma and chroma data without increasing overall bandwidth, resolving the contradiction between maintaining picture quality and reducing bandwidth requirements.

Inventive Principle:
Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent combines luma and chroma sample transmission by packing both types of samples into the same data stream using a unified 16-bit word format. Instead of separate transmission channels, luma and chroma data are merged into compacted packets where each packet contains both luma and chroma information in a space-efficient manner, reducing total bandwidth consumption while preserving 10-bit quality for both components.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of operation

If traditional byte-aligned formatting is used for simplicity, then ease of operation is maintained, but processing efficiency decreases for 10-bit samples

Engineering Contradiction:
Improveformatting simplicityVSAvoidprocessing efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent changes the fundamental data organization parameter from traditional byte-aligned 8-bit sampling to 10-bit sampling within 16-bit words. This parameter change enables more efficient use of available bit space, allowing 6 unused bits per luma sample to be repurposed for chroma data storage. The new formatting scheme optimizes data density and processing efficiency specifically for 10-bit 4K video applications.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent utilizes the previously unused upper bits (6 bits) of 16-bit word containers as an additional storage dimension. Instead of treating each 16-bit word as containing only a single 10-bit luma sample with wasted space, the invention adds a second data layer by embedding chroma sample portions in the upper bits, effectively creating a two-layer data structure that increases information density without requiring additional transmission bandwidth.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10306200B2Efficient 10-bit pixel formatting of uncompressed 4K video signals for distributed processing
Publication Date: 2019.05.28 CISCO TECHNOLOGY INC
  • US10306200B2 patent drawing
  • US10306200B2 patent drawing
  • US10306200B2 patent drawing

AI summary

A method including: receiving 16 10-bit Y samples stored in respective 16-bit words; logically ORing the 16 10-bit Y samples with the lowest 5-bits corresponding to each of the next 16 10-bit U samples; receiving the first consecutive 16 10-bit Y samples stored in respective 16-bit words; logically ORing the next consecutive 16 10-bit Y samples with the highest 5-bits corresponding to each of the next 16 10-bit U samples; receiving the second consecutive 16 10-bit Y samples stored in respective 16-bit words, logically ORing the second consecutive 16 10-bit Y samples with the lowest 5-bits corresponding to each of the next 16 10-bit V samples; receiving the third consecutive 16 10-bit Y samples stored in respective 16-bit words; and logically ORing the third consecutive 16 10-bit Y samples with the highest 5-bits corresponding to each of the next 16 10-bit V samples.