Dynamic Digital Predistortion for GaN Amplifier Trapping Effects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Power amplifiers, particularly those using Gallium Nitride (GaN) technology, experience non-linearity due to electron trapping and de-trapping effects, leading to distortion in output signals, which existing digital pre-distortion techniques struggle to fully compensate for.
Innovation Solution
The implementation of dynamic digital pre-distortion correction circuitry that generates long-term signal statistics and nonlinear terms to adjust input data signals, combining these to mitigate non-linearity introduced by GaN power amplifiers, using techniques like Generalized Memory Polynomial models and lookup tables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If dynamic digital pre-distortion correction is implemented to compensate for non-linearity in GaN power amplifiers, then signal linearity and transmission quality are improved, but device complexity and computational requirements increase
Solution Approach 1:
The pre-distortion correction is divided into two separate circuitries: one for fast dynamics terms and one for slow dynamics terms. This segmentation allows each circuitry to be optimized independently for its specific time constant range, reducing the overall complexity compared to a single comprehensive circuitry while maintaining signal linearity across different dynamics.
Solution Approach 2:
The system dynamically adapts by generating different predistortion terms based on the time constant characteristics of the input signal statistics. Fast dynamics terms handle rapid signal variations while slow dynamics terms handle gradual variations, allowing the system to maintain optimal linearity compensation across varying operating conditions without requiring a fixed complex structure.
2Measurement precision
If long-term signal statistics are generated and processed to account for slow dynamics, then compensation accuracy is improved, but processing time and computational load increase
Solution Approach 1:
Signal statistics are processed through separate pathways for fast and slow dynamics terms. The slow dynamics pathway processes long-term statistics at reduced rates, while fast dynamics pathway handles short-term statistics at higher rates. This segmentation allows accurate long-term compensation without requiring all processing to occur at maximum speed, reducing overall processing time while maintaining accuracy.
Solution Approach 2:
The system updates predistortion terms periodically based on the characteristics of the input signal rather than continuously processing all signal variations. By identifying and processing only the relevant time constant ranges, the system achieves accurate long-term compensation at periodic intervals rather than requiring continuous full-speed processing.
Data Source
AI summary
Methods, apparatus, systems, and articles of manufacture are described for dynamic digital pre-distortion correction. An example system includes programmable circuitry operable to execute computer readable instructions to at least: generate signal statistics based on an input signal; group the signal statistics into a first group of signal statistics or a second group of signal statistics based on time constants of the signal statistics; decimate the first group of signal statistics; generate a first predistortion term based on the decimated first group of signal statistics; generate a second predistortion term based on the second group of signal statistics; and generate an output predistortion terminal based on the first predistortion term and the second predistortion term.