Baseband Predistortion Using Pre-Equalized LUTs for Wideband Linearity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing baseband predistortion methods struggle to effectively compensate for both RF frequency response memory effects and bias-induced or thermal memory effects in multi-channel wideband wireless transmitters, leading to limitations in linearity and efficiency due to high computational complexity and numerical instability.
Innovation Solution
A piecewise pre-equalized lookup table (LUT) based digital predistortion system is employed, which reduces computational complexity and numerical instability while compensating for electrical and thermal memory effects, using a LUT with pre-equalized coefficients to address memory effects and improve linearity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Volterra filter structure is used to compensate memory effects, then memory effects compensation is improved, but device complexity increases significantly
Solution Approach 1:
The patent segments the memory effects compensation into two separate components: a FIR filter for short-term memory effects and a lookup table for long-term memory effects. This segmentation allows each component to handle specific types of memory effects independently, reducing the overall computational complexity compared to using a full Volterra filter while maintaining effective compensation.
Solution Approach 2:
The patent extracts and separates the memory effects compensation from the nonlinearity compensation, handling them through different mechanisms. The FIR filter extracts short-term memory effects, while the lookup table handles long-term memory effects, allowing each to be optimized independently rather than requiring a complex unified approach.
2Device complexity
If memory polynomial structure is used to reduce coefficients, then device complexity is reduced, but manufacturing precision decreases due to numerical instability
Solution Approach 1:
The patent uses a lookup table as a simpler, more stable alternative to memory polynomials for handling long-term memory effects. While lookup tables require memory storage, they avoid the numerical instability issues of polynomial coefficient estimation, providing a more reliable solution that can be pre-computed and stored.
3Device complexity
If Hammerstein predistorter is used to reduce complexity, then device complexity is reduced, but reliability decreases due to inability to compensate certain memory effects
Solution Approach 1:
The patent merges the Hammerstein predistorter (FIR filter for short-term memory) with a lookup table (for long-term memory effects) to create a more comprehensive solution. This combination allows the system to compensate for both types of memory effects while maintaining relatively low computational complexity compared to full Volterra or memory polynomial approaches.
4Reliability
If digital predistortion is applied to improve linearity, then linearity is improved, but use of energy increases due to computational requirements
Solution Approach 1:
The patent segments the predistortion processing into two stages: a FIR filter for short-term memory effects and a lookup table for long-term memory effects. This segmentation allows the system to achieve effective linearization while reducing computational energy consumption by using simpler operations (filtering and table lookup) rather than complex polynomial calculations.
Data Source
AI summary
In some embodiments, a method of reducing adjacent channel power ratio and compensating memory effects of multi-channel wideband communication systems using multiplexing modulation techniques is provided. The method includes generating an address from samples of a baseband input signal of a communication system. The method also includes retrieving from a memoryless lookup table an entry in accordance with the address. The method further includes pre-equalizing the baseband input signal, the pre-equalization depending on one or more magnitudes of the input signal. The method still further includes multiplying the pre-equalized baseband input signal and the lookup table entry.