TCM Decoder Region Decision for High-Order QAM Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of TCM decoders increases significantly with higher QAM modulation orders, requiring extensive calculations of Euclidean distances between received signals and constellation points, which hinders efficient decoding.
Innovation Solution
A decoding method that reduces complexity by performing region decision on the received signal using S region decision formulas, dividing constellation points into N subsets, and acquiring branch metrics and non-encoded bits from the closest constellation points, allowing for Viterbi decoding without calculating Euclidean distances from all constellation points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Euclidean distances between received signal and all constellation points are calculated to determine closest constellation points, then decoding accuracy is improved, but decoder complexity increases significantly
Solution Approach 1:
The patent segments the constellation diagram into multiple regions using decision formulas, where each region contains one or more constellation points. Instead of calculating distances to all constellation points, the decoder only needs to identify which region the received signal falls into and select the corresponding constellation point(s) in that region, dramatically reducing the number of distance calculations required while maintaining decoding accuracy.
2Productivity
If QAM modulation order is increased to achieve higher spectral efficiency, then transmission capacity is improved, but the number of constellation points increases causing decoder complexity to increase proportionally
Solution Approach 1:
The patent divides the large number of constellation points into multiple regions using decision formulas. For high-order QAM modulation with many constellation points, this segmentation approach allows the decoder to work with a reduced set of candidate points within each region rather than all constellation points, keeping decoder complexity manageable while supporting high spectral efficiency.
Solution Approach 2:
The patent pre-establishes decision formulas and region boundaries during the encoding phase. These formulas are designed beforehand to partition the constellation diagram in a way that facilitates efficient decoding. By preparing the region structure in advance, the decoder avoids the need for complex real-time calculations when handling high-order modulation schemes.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present invention relate to the field of communications, and provide a decoding method and a decoder, which are used to reduce decoding complexity. The method includes: receiving a to-be-decoded signal; performing region decision on the to-be-decoded signal according to a region decision rule formed by S region decision formulas, to acquire a region decision result; acquiring N constellation points according to the decision result, where the N constellation points are separately constellation points that are in the N subsets and that are closest to the to-be-decoded signal; acquiring N non-encoded bits corresponding to the N constellation points, and branch metrics between the to-be-decoded signal and the N constellation points; and performing Viterbi decoding based on the branch metrics and the N non-encoded bits, and outputting a decoding result corresponding to the to-be-decoded signal. The present invention is applicable to a signal decoding scenario.