Cascaded Polar-Convolutional Encoding for Better Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The decoding performance of existing polar encoding methods needs to be improved to enhance data transmission reliability in communication systems.

Innovation Solution

A cascaded encoding method is introduced, where polar encoding is followed by convolutional encoding using a feedforward shift register, with specific conditions such as odd number of taps and initial bits being the last m bits of the sequence, to improve decoding performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If polar encoding is used alone, then encoding complexity is low, but decoding performance is insufficient

Engineering Contradiction:
Improvedecoding performanceVSAvoidencoding complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines polar encoding and convolutional encoding into a cascaded encoding structure. The polar encoder first processes the input bit sequence to generate an intermediate bit sequence, which is then fed into a convolutional encoder. This merging of two encoding schemes allows the system to maintain the low complexity advantage of polar encoding while adding the error correction capabilities of convolutional encoding, thereby improving overall decoding performance without excessively increasing system complexity.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a composite encoding scheme by integrating polar code and convolutional code. The polar code component provides polarization properties for reliable bit channel construction, while the convolutional code component adds redundancy and error correction capability. This composite approach leverages the strengths of both coding theories to achieve better decoding performance than either scheme could provide alone, while keeping the overall structure manageable through systematic design.

Inventive Principle:
Principle #40Composite materials

2Reliability

If convolutional encoding is added after polar encoding, then decoding performance is improved, but encoding complexity increases

Engineering Contradiction:
Improvedecoding performanceVSAvoidencoding complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the encoding process into two distinct stages: a polar encoding stage that processes the input bit sequence to generate an intermediate representation, and a convolutional encoding stage that further processes this intermediate sequence. This segmentation allows each encoder to be optimized independently and enables modular implementation, where the convolutional encoder is added as a separate functional block after the polar encoder, making the complexity increase manageable and systematic.

Inventive Principle:
Principle #1Segmentation

3Reliability

If tail-biting convolutional encoding is used, then decoding reliability is improved, but implementation complexity increases

Engineering Contradiction:
Improvedecoding reliabilityVSAvoidimplementation complexity
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent implements tail-biting convolutional encoding by pre-configuring the shift register with specific initial states that match the final states of the encoding sequence. This preliminary setup of the shift register ensures that the encoding process starts and ends in a consistent state, eliminating the need for additional termination bits or complex state management during decoding. The initial state configuration is predetermined based on the encoding parameters, simplifying the overall implementation while maintaining the reliability benefits of tail-biting structures.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4109766B1Encoding method and device
Publication Date: 2025.10.01 HUAWEI TECH CO LTD
  • EP4109766B1 patent drawingFigure 1~2
  • EP4109766B1 patent drawingFigure 3~4
  • EP4109766B1 patent drawingFigure 5~7A

AI summary

An encoding method and apparatus are provided. The method includes: obtaining a to-be-encoded bit sequence (S401); performing polar encoding on the to-be-encoded bit sequence to obtain a first bit sequence (S402); performing convolutional encoding on the first bit sequence to obtain a second bit sequence (S403); and outputting the second bit sequence (S404). In the foregoing process, because convolutional encoding is cascaded after polar encoding, an encoding process of the to-be-encoded bit sequence is more adequate, and decoding performance can be improved.