QAM Demapper Adaptive Offset Correction for Non-Linear Distortion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-order Quadrature Amplitude Modulation (QAM) systems in digital communication face challenges with noise and non-linear distortions, leading to reduced receiver sensitivity and increased power requirements due to random noise and deterministic impairments in the transmission channel.
Innovation Solution
The proposed method involves adaptive demapping with soft output, where the input data is sliced into sectors to determine a subset of signal points, and per-symbol offsets are calculated to correct for non-linear distortions, enabling improved Log-likelihood Ratio (LLR) computation and error correction in the receiver.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If higher order QAM modulation is used to increase data capacity, then optical channel capacity is improved, but receiver sensitivity deteriorates due to noise and non-linear distortions
Solution Approach 1:
The patent divides the QAM constellation into multiple sectors, each processed independently with dedicated offset correction. This segmentation allows the receiver to handle non-linear distortions more effectively by applying sector-specific corrections, thereby maintaining reliability while supporting higher order modulations for increased capacity.
Solution Approach 2:
The patent dynamically adjusts per-symbol offsets to compensate for non-linear distortions in the transmission channel. By changing the offset parameters adaptively based on received signal characteristics, the system maintains accurate symbol detection despite increasing modulation complexity, thus preserving receiver sensitivity while enabling higher data capacities.
2Reliability
If per-symbol offset correction is applied to correct non-linear distortions, then error performance is improved, but computational complexity increases
Solution Approach 1:
The patent limits offset correction calculations to only those symbols within the determined sector, rather than processing the entire constellation. This segmented approach reduces computational complexity by focusing resources on relevant symbols while still achieving effective distortion correction and improved error performance.
Solution Approach 2:
The patent applies offset correction selectively to a subset of symbols that are most relevant to the received signal, rather than uniformly processing all possible symbols. This partial action approach achieves sufficient error performance improvement without the full computational burden of exhaustive symbol-by-symbol correction.
3Reliability
If adaptive demapping with sector slicing is used to determine subset of signal points, then receiver sensitivity is enhanced, but processing complexity increases
Solution Approach 1:
The patent divides the constellation into sectors and performs demapping independently within each sector. This segmentation simplifies the demapping process by reducing the search space from the entire constellation to only relevant sectors, thereby enhancing receiver sensitivity while keeping processing complexity manageable through localized processing.
Data Source
AI summary
Techniques are presented for receiving Quadrature Amplitude Modulated (QAM) symbols from a transmitter via a transmission path. In one example, a demodulator is configured to down-convert an incoming Radio Frequency (RF) signal to a baseband signal and convert the baseband signal to digital samples, and output the digital samples. A demapper is configured to receive the digital samples output from the demodulator and output data encoded in QAM symbols. The demapper is further configured to: determine from a constellation of QAM symbols a subset of QAM symbols that a digital sample from the demodulator may represent; apply an offset to each QAM symbol in the subset of QAM symbols of the constellation to result in a subset of offset QAM symbols; determine which QAM symbol in the subset of offset QAM symbols the digital sample most likely represents; and output data representing a determined QAM symbol.