Dual-Band Digital Predistortion for Wideband Power Amplifier Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital pre-distortion techniques face challenges in accurately mitigating distortion caused by power amplifiers driven into compression, especially as wireless communication systems transition to wider bandwidths, limiting their effectiveness in modern wireless communication systems.
Innovation Solution
A dual-band digital pre-distortion circuitry that models a power amplifier using a binomial expansion-based model structure with optimized delay combinations, allowing for increased model accuracy and flexibility in selecting delays to adapt to wider bandwidths, comprising complex sub-band signals and their conjugates with specific delays, and applying these optimized combinations to improve pre-distortion performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional digital pre-distortion techniques are used, then implementation is simpler, but model accuracy deteriorates when adapting to wider bandwidths
Solution Approach 1:
The patent segments the baseband signal into multiple sub-band signals (first complex sub-band signal and second complex sub-band signal) corresponding to different frequency bands. This segmentation allows the model to process each band separately with appropriate delay combinations, improving accuracy for wideband signals while managing complexity through modular processing of individual bands rather than treating the entire bandwidth as a single unit.
Solution Approach 2:
The patent introduces delay combinations as an additional dimension parameter beyond the traditional single-delay approach. By optimizing multiple delay values (first delay, second delay, third delay, fourth delay, fifth delay) for different signal components, the model gains extra degrees of freedom to accurately represent distortion characteristics across wide bandwidths, effectively adding a temporal dimension to the pre-distortion modeling.
2Measurement precision
If more delays are applied to increase model accuracy, then distortion mitigation improves, but optimization time increases
Solution Approach 1:
The patent divides the optimization problem into separate sub-problems for each sub-band signal. By optimizing delay combinations for the first complex sub-band signal and second complex sub-band signal independently, the overall optimization complexity is reduced compared to optimizing all delays simultaneously for the entire bandwidth, thereby decreasing optimization time while maintaining accuracy.
Solution Approach 2:
The patent applies a structured subset of delay combinations rather than exhaustively optimizing all possible delay parameters. The model structure specifies particular delay relationships (first delay for direct term, second and third delays for conjugate multiplication terms, fourth and fifth delays for the other sub-band terms), which provides sufficient accuracy for wideband operation without requiring complete optimization of every possible parameter combination, thus reducing optimization time.
3Adaptability or versatility
If fixed delay structures are used, then model implementation is simpler, but adaptability to wider bandwidths deteriorates
Solution Approach 1:
The patent implements dynamic delay selection where the delay values (first delay, second delay, third delay, fourth delay, fifth delay) are optimized based on the specific operating conditions and bandwidth requirements. This dynamic approach allows the model to adapt to different bandwidth scenarios by adjusting the delay parameters, rather than being constrained by fixed delay structures, thereby improving versatility across wideband applications.
Solution Approach 2:
The patent creates a universal model structure that can handle multiple frequency bands and bandwidth configurations through the same architectural framework. The sub-band decomposition approach with optimized delay combinations serves as a multi-functional solution that adapts to different bandwidth requirements (wideband, ultra-wideband) without requiring fundamentally different model structures, achieving versatility through a unified adaptable framework.
Data Source
AI summary
A model structure modeling a power amplifier is based on at least a binomial expansion, a first building block, a second building block, and a third building block. The first building block is a first complex sub-band signal with a first delay, the second building block is a multiplication of the first complex sub-band signal with a second delay and a complex conjugate of the first complex sub-band signal with a third delay, and the third building block is a multiplication of a second complex sub-band signal with a fourth delay and a complex conjugate of the second complex sub-band signal with a fifth delay. The sum of the first complex sub-band signal and the second complex sub-band signal is a baseband signal. Terms are obtained by optimizing delay combinations for the model structure. The model structure is used to dual-band digital pre-distortion of the baseband signal.