LDPC H-Matrix Swapping for Balanced Parallel Decoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing LDPC decoders experience delays due to uneven processing times between zero and non-zero circulants, as one decoding logic unit must wait for the other to finish processing before retrieving the next block, leading to inefficiencies in decoding.

Innovation Solution

A modified LDPC H matrix is used, where circulants in one column are swapped with those in another, allowing for even distribution of non-zero circulants across decoding logic units, enabling parallel processing without imbalances and minimizing delays.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If parallel decoding using two decoding logic units is implemented, then decoding throughput is improved, but processing delays occur due to uneven processing times between zero and non-zero circulants

Engineering Contradiction:

Improvedecoding throughputVSAvoidprocessing delay

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent modifies the H matrix by swapping circulants between columns to change the processing parameters. Specifically, non-zero circulants are redistributed across different columns so that each decoding logic unit receives a balanced mix of zero and non-zero circulant processing tasks, equalizing their processing times and eliminating idle waiting periods

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces dynamic task allocation by modifying the H matrix structure to allow flexible assignment of decoding tasks. The swapped circulant configuration enables the system to dynamically balance the workload between decoding logic units, ensuring that both units remain actively engaged throughout the decoding process without idle waits

Inventive Principle:

Principle #15Dynamics

2Speed

If decoding logic units process blocks in parallel, then decoding speed is improved, but one unit must wait for the other to finish processing blocks from buffers

Engineering Contradiction:

Improvedecoding speedVSAvoidprocessing continuity

Core Design Contradiction:

SpeedVSEase of operation

Solution Approach 1:

The H matrix is modified by swapping circulants to change the processing parameters of each decoding logic unit. This ensures that both units have comparable processing times for their assigned blocks, allowing them to complete processing simultaneously and retrieve the next blocks from buffers without one unit waiting for the other

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent performs preliminary modification of the H matrix before decoding begins. By pre-swapping circulants to create a balanced distribution, the system prepares the decoding tasks in advance so that parallel processing can proceed continuously without idle waits, maintaining both high speed and operational smoothness

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10419026B2Method and apparatus for efficient data decoding

Publication Date: 2019.09.17 CHENGDU XINSHENG INTEGRATED CIRCUIT CO LTD

AI summary

A method and apparatus for efficient data decoding is described. Data is encoded by an LDPC encoder using a G matrix. An LDPC decoder uses a modified H matrix to decode encoded blocks of data, the modified H matrix having at least two columns of its circulants swapped with each other. The encoded blocks of data are stored, decoded and reconstructed in an order that considers the circulants in the columns that have been swapped.