Tensor Product Coding With Iterative Parity Hiding for Short Blocks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Linear block codes with shorter input block lengths, used in data recording and communication, face a tradeoff between coding gain and code rate, requiring higher overhead and reducing performance.

Innovation Solution

The use of tensor product codes, specifically combining an inner code and an outer iterative parity hiding code, such as Turbo or LDPC codes, to encode and decode data streams, allowing for shorter block lengths without the full code rate penalty, through iterative decoding and reliability metric refinement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If linear block codes with shorter input block lengths are used, then coding gain is improved, but code rate penalty increases

Engineering Contradiction:

Improvecoding gainVSAvoidcode rate

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The code is segmented into an inner code and an outer iterative code, where the inner code handles shorter block lengths for coding gain while the outer code structure manages the overall code rate, preventing the full penalty associated with short block lengths

Inventive Principle:

Principle #1Segmentation

2Reliability

If linear block codes with shorter input block lengths are used, then error detection and correction capability is improved, but overhead increases

Engineering Contradiction:

Improveerror detection and correction capabilityVSAvoidoverhead

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines an inner code with an outer iterative parity hiding code to create a tensor product code structure that achieves strong error detection and correction capabilities while reducing the overhead typically associated with short block length codes

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS8086945B1Tensor product codes containing an iterative code

Publication Date: 2011.12.27 MARVELL ASIA PTE LTD

AI summary

Systems and methods are provided for encoding a stream of datawords based on a tensor product code to provide a stream of codewords, and detecting and decoding a stream of received data based on a tensor product code to provide a decoded stream of data. In one aspect, the tensor product code is based on two codes including an inner code and an outer parity hiding code, where the outer parity hiding code is an iterative code. In certain embodiments, the outer parity hiding code is a Turbo code or a low density parity check (LDPC) code.