Iterative Receiver Detection With Decoder Feedback for Lower BER

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication systems face challenges in accurately estimating data transmitted by a transmitter, particularly when sub-optimal bit labelling leads to reduced bit error rate (BER) performance.

Innovation Solution

Implementing a receiver algorithm with trainable parameters that operates on channel outputs and feedback from an outer channel decoder, utilizing machine learning principles for iterative demapping and decoding to improve BER performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If sub-optimal bit labelling is used in the communication system, then the device complexity is reduced, but the bit error rate performance deteriorates

Engineering Contradiction:
Improvebit labelling complexityVSAvoidbit error rate performance
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent implements an iterative detection framework where the receiver algorithm receives feedback from the channel decoder about the detected data. This feedback loop allows the receiver to refine its estimates of the transmitted data by incorporating information from both the channel output and the decoder's previous detection results, thereby improving BER performance without changing the bit labelling scheme.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The receiver algorithm employs trainable parameters that are optimized through machine learning techniques. By adjusting these parameters based on training data and iterative refinement, the system adapts to compensate for the sub-optimal bit labelling, improving detection accuracy while maintaining the simplicity of the original labelling scheme.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If iterative demapping and decoding with trainable parameters is implemented, then the bit error rate performance is improved, but the device complexity increases

Engineering Contradiction:
Improvebit error rate performanceVSAvoidreceiver algorithm complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces traditional deterministic receiver algorithms with a machine learning-based approach using trainable parameters. This substitution allows the system to learn optimal detection strategies from data, achieving superior BER performance. The trainable parameters are optimized using gradient-based methods, replacing conventional signal processing techniques.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The receiver algorithm undergoes a training phase before actual operation, where it learns optimal parameter values from training data. This preliminary action prepares the system for improved performance during operational phase, allowing the iterative detection to converge faster and achieve better BER performance with fewer iterations during actual use.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If machine learning principles are used for receiver optimization, then the bit error rate performance is enhanced, but the training time and computational resources increase

Engineering Contradiction:
Improvebit error rate performanceVSAvoidtraining time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements a practical compromise where the training process is performed to a sufficient level rather than exhaustive optimization. The receiver is trained on representative training data to achieve adequate performance, and then deployed for operational use. This partial action approach balances training time investment with operational performance gains, avoiding unnecessary excessive training that would yield diminishing returns.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3963765B1Iterative detection in a communication system
Publication Date: 2026.01.28 NOKIA TECHNOLOGIES OY
  • EP3963765B1 patent drawingFigure 1~2
  • EP3963765B1 patent drawingFigure 3~4
  • EP3963765B1 patent drawingFigure 5~6

AI summary

An apparatus, computer program and method is described comprising receiving data at a receiver of a communication system, generating an estimate of the data as transmitted by a transmitter of the transmission system (wherein generating the estimate comprises a receiver algorithm having at least some trainable weights), generating a refined estimate of the transmitted data, based on said estimate and an error correction algorithm (wherein, in an operational mode, said estimate of the data as transmitted is generated based on the received data and said refined estimate); and generating, in the operational mode, a revised estimate of the transmitted data on each of a plurality of iterations of said generating an estimate of the transmitted data until a first condition is reached.