Beamforming PA Linearization Using Fixed DPD and Neural Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Massive multiple input multiple output (mMIMO) systems face challenges with computationally intensive and costly digital predistortion (DPD) for power amplifiers (PAs) in wireless communication, particularly in 5G technology, which is complicated and requires individual feedback for each PA.
Innovation Solution
Implementing a fixed digital predistortion (DPD) circuit with an analog neural network (NN) to control power amplifiers (PAs) to mimic an idealized amplification curve, reducing the need for feedback and computational intensity, and integrating edge computing for training and parameter updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If digital predistortion (DPD) is implemented for each power amplifier in mMIMO systems, then linearity and performance are improved, but computational complexity and system cost increase significantly
Solution Approach 1:
The patent merges the DPD functions across multiple power amplifiers into a shared computational resource. Instead of implementing independent DPD for each PA, a single DPD engine processes signals for multiple PAs by exploiting their correlated non-linear characteristics, thereby reducing overall computational complexity while maintaining linearity performance.
Solution Approach 2:
The DPD system is designed with universal functionality to serve multiple power amplifiers simultaneously. A single DPD implementation can be applied across different PAs in the mMIMO system, making the complex computational resource universal rather than dedicated to each individual amplifier.
2Measurement precision
If individual feedback loops are implemented for each power amplifier, then accurate linearization is achieved, but system cost and complexity increase
Solution Approach 1:
The patent merges individual feedback loops into a shared feedback mechanism. Multiple PAs contribute to a common feedback signal that is processed by a single DPD engine, eliminating the need for separate feedback paths for each amplifier while preserving linearization accuracy through the exploitation of signal correlations.
3Reliability
If full adaptive DPD is implemented for each power amplifier, then optimal performance is achieved, but computational intensity and processing time increase
Solution Approach 1:
The patent merges adaptive DPD computations across multiple PAs into a unified processing framework. By combining the adaptation processes and exploiting the fact that multiple PAs experience similar operating conditions, the system achieves optimal performance with reduced computational intensity and faster processing times compared to individual adaptive DPD implementations.
Data Source
AI summary
Apparatus for causing a power amplifier (PA) to act as an idealized power amplifier (iPA). The apparatus comprises: a fixed digital predistortion (DPD) circuit; a PA coupled to the DPD circuit; at least one sensor adapted to determine at least one condition related to of the power amplifier; and a neural network adapted control a level of amplification provided by the power amplifier based on a measurement of the at least one condition as measured by the at least one sensor.