Sequence Detection in High-Order QAM Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-order quadrature amplitude modulation (QAM) systems face inefficiencies in maximum likelihood sequence estimation (MLSE) due to exponentially increasing operations, leading to high complexity and low detection efficiency in existing algorithms.
Innovation Solution
A method that determines the maximum possible candidate values for each element in a received sequence, reserves state sequences corresponding to these values, and performs likelihood computation only on the most consistent reserved sequences to detect the sent sequence, thereby reducing unnecessary computations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If maximum likelihood sequence estimation (MLSE) with Viterbi algorithm is used to determine sent sequence, then detection accuracy is improved, but computational complexity increases exponentially with system order
Solution Approach 1:
The patent segments the computational process into two distinct stages: first performing hard decision on received symbols to obtain initial candidate values, then performing likelihood computation only on sequences formed from these candidates. This segmentation divides the originally exponential search space into a manageable two-step process, reducing computational complexity while preserving detection accuracy through the subsequent likelihood verification step.
Solution Approach 2:
The patent performs preliminary hard decision on received symbols before the full likelihood computation. By making initial decisions on each received symbol to determine candidate values, the system prepares a reduced set of plausible sequences in advance, which then undergoes likelihood computation. This preliminary action eliminates the need to evaluate all possible sequences, thereby reducing computational complexity exponentially.
2Productivity
If system order in quadrature amplitude modulation is increased to improve data rate, then productivity is improved, but the quantity of operations in MLSE-VA algorithm increases exponentially
Solution Approach 1:
The patent segments the sequence detection into symbol-level hard decision followed by sequence-level likelihood computation. This allows high-order QAM systems to first determine candidate symbols independently, then evaluate sequences formed from these candidates, reducing the operational burden that would otherwise grow exponentially with system order.
Solution Approach 2:
The patent performs partial likelihood computation only on sequences formed from hard decision candidates rather than all possible sequences. This partial action approach computes likelihood for a subset of sequences that are most likely to be correct, achieving sufficient detection accuracy without the exponential operational cost of evaluating all possible sequences in high-order QAM.
Data Source
AI summary
A method for detecting a sent sequence, a receiver, and a receiving device in order to simplify an algorithm for detecting a sent sequence and improve detection efficiency. The method for detecting a sent sequence includes determining a maximum possible candidate value of each element in N elements of a received element sequence to obtain N maximum possible candidate values, where N is a positive integer, determining state sequences corresponding to the N maximum possible candidate values as reserved sequences to obtain N groups of reserved sequences, performing likelihood computation on the N groups of reserved sequences, and setting a reserved sequence that is in the N groups of reserved sequences and is most consistent with the element sequence as a detected sent sequence.