Iterative Signal Decoding With Partial Detector Updates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In communication systems, particularly with Bit Interleaved Coded Modulation (BICM), the performance penalty due to inter-symbol interference, phase ambiguities, and MIMO transmissions is not fully recovered with a limited number of iterations in practical receivers, as the computationally complex demodulator/differential decoder/MIMO detector execution is restricted.

Innovation Solution

The method involves iterative decoding with a detector and decoder exchanging extrinsic information, where the detector operates partially on subsets of the received signal, updating reliability information selectively in each iteration, and using a loop detector to provide updated a-priori information for subsequent decoding, allowing for faster convergence and reduced complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If the demodulator/differential decoder/MIMO detector operates with a limited number of iterations in practical receivers, then the implementation complexity is reduced, but the performance penalty due to inter-symbol interference, phase ambiguities, and MIMO transmissions is not fully recovered

Engineering Contradiction:
Improvereceiver complexityVSAvoidperformance penalty recovery
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The received signal is divided into multiple parts or segments, and the detector selectively updates reliability information for different parts in different iterations. This segmentation allows the system to achieve performance recovery with fewer full-signal iterations, thereby reducing receiver complexity while maintaining reliability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of performing complete detection operations on the entire received signal in each iteration, the detector performs partial detection by selectively updating reliability information for specific parts of the signal. This partial action reduces computational complexity while still achieving adequate performance penalty recovery through iterative refinement

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If the detector updates reliability information for the whole received signal in every iteration, then the BER convergence is improved, but the computational complexity of the receiver increases

Engineering Contradiction:
ImproveBER convergenceVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The received signal is divided into multiple parts, and the detector updates reliability information selectively for different parts in different iterations. This segmentation maintains BER convergence by ensuring all parts are eventually updated while reducing computational complexity by avoiding redundant full-signal processing in every iteration

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The detector employs periodic updating patterns where different parts of the received signal are updated in an alternating or rotating manner across iterations. This periodic action ensures that all parts receive updates over time (maintaining convergence) while distributing the computational load (reducing per-iteration complexity)

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP2860895B1Apparatus and method for Iteratively decoding a received signal
Publication Date: 2016.09.14 ALCATEL LUCENT SA
  • EP2860895B1 patent drawingFigure 1a~1b
  • EP2860895B1 patent drawingFigure 1c~2c
  • EP2860895B1 patent drawingFigure 3

AI summary

Embodiments relate to a concept for iteratively decoding a received signal (z; 401) comprising encoded information symbols. The concept includes detecting (402) estimates (403) of encoded information symbols corresponding to the received signal (401) to obtain a-priori information for a decoder (404). An iteration (l) includes generating (404), as extrinsic information (405), estimates for decoded information symbols corresponding to at least a part (z[p,1]; z[p,2]) of the received signal (z; 401) using the a-priori information (403) and the decoder (404), updating (406) estimates (403-l) of encoded information symbols corresponding to the part (z1; z2) of the received signal (z; 401) using the received signal (z; 401) and the extrinsic information (405-(l-1)) corresponding to the part (z[p,1]; z[p,2]) of the received signal, and providing the updated estimates (403-l) to the decoder (404) as updated a-priori information for a subsequent iteration.