Cascaded Lookup Table Generation for Non-Linear Signal Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The resources required for a lookup table (LUT) that stores non-linear deviation values of N symbol sequences increase exponentially with N, leading to significant resource usage, especially as the modulation mode and levels of the electrical input signal increase.
Innovation Solution
Divide the N symbol sequence into two sub-sequences, with statistical averaging and non-linear deviation calculation for each sub-sequence, forming a cascaded lookup table, and optionally reduce the number of levels for edge symbols, allowing for a more efficient storage and processing of non-linear deviation values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the LUT stores non-linear deviation values for N symbol sequences to achieve better compensation effectiveness, then the compensation effectiveness is improved, but the LUT resources increase exponentially
Solution Approach 1:
The patent divides the N symbol sequence into two sub-sequences of length N/2, and separately calculates non-linear deviation values for each sub-sequence. This segmentation allows the LUT to store only N/2 entries instead of N entries, reducing the LUT resources by half while maintaining compensation effectiveness through the combined use of both sub-sequence deviations.
Solution Approach 2:
The patent extracts the non-linear deviation calculation into separate operations for each sub-sequence. By calculating deviations for the first sub-sequence (symbols 1 to N/2) and the second sub-sequence (symbols N/2+1 to N) independently, the system reduces the computational burden and LUT size while preserving the essential compensation information.
2Reliability
If the LUT stores non-linear deviation values for longer symbol sequences to improve compensation, then the compensation effectiveness is improved, but the hardware resources required increase significantly
Solution Approach 1:
The patent segments the symbol sequence into two equal parts and processes them independently. This segmentation strategy reduces the LUT size from 2^N entries to 2^(N/2) entries, significantly reducing hardware resources while maintaining the ability to compensate for non-linear effects in the original N-length symbol sequence.
Solution Approach 2:
The patent transforms the problem from a one-dimensional LUT of size 2^N to a two-dimensional approach with two LUTs of size 2^(N/2). This dimensional change allows the system to achieve the same compensation effectiveness with reduced hardware resources by distributing the lookup operations across two smaller tables.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
Embodiments of the present invention provide a lookup table generation method and apparatus and a pre-compensation method and apparatus, so as to resolve a current problem that resources of an LUT increase exponentially with an increase in a length of a symbol sequence of the LUT. The method includes: dividing a first symbol sequence into two sub-symbol sequences, where symbols that are from the first symbol to a center symbol in the first symbol sequence form one sub-symbol sequence, and symbols that are from the center symbol to the last symbol in the first symbol sequence form the other sub-symbol sequence; performing statistical averaging on each sub-symbol sequence, and calculating a non-linear deviation value of the sub-symbol sequence according to a result of statistical averaging; correspondingly storing a sub-symbol sequence in which the center symbol is the last symbol and a non-linear deviation value of the sub-symbol sequence to form a first lookup table in a cascaded lookup table; and correspondingly storing a sub-symbol sequence in which the center symbol is the first symbol and a non-linear deviation value of the sub-symbol sequence to form a second lookup table in the cascaded lookup table.