Digital Pre-Distortion Auto-Tuning for Doherty Power Amplifiers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional analog Doherty power amplifiers suffer from inefficiencies and limited operational bandwidth due to fixed configurations and the need for manual tuning, which is not adaptable to varying input conditions and circuit states.

Innovation Solution

A digital Doherty power amplifier system with a learning-based auto-tuning method that adaptively optimizes control parameters and learning cost functions to enhance efficiency and linearity, using a model-free algorithm for black-box optimization and data-driven optimization techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional analog DPA with fixed configuration is used, then device complexity is reduced, but adaptability to varying input conditions deteriorates

Engineering Contradiction:
Improveadaptability to varying input conditionsVSAvoiddevice complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic adaptability by introducing an autotuning controller that continuously adjusts control parameters based on real-time circuit states and input signals. The system transitions from fixed analog configuration to dynamic digital control, enabling adaptation to varying input conditions while maintaining manageable complexity through automated tuning algorithms.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes physical and operational parameters dynamically through digital control. The autotuning controller modifies control parameters such as phase alignment, power splitting ratio, and amplifier bias points based on measured circuit states, enabling the DPA to adapt to different operating conditions without requiring physical reconfiguration.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If manual tuning is performed to optimize PA performance, then manufacturing precision is improved, but productivity deteriorates

Engineering Contradiction:
Improveoptimal control parametersVSAvoidtuning process efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent implements self-service through the autotuning controller that automatically measures circuit states, evaluates performance metrics, and adjusts control parameters without human intervention. The system performs its own optimization by comparing actual performance against target specifications and making real-time adjustments, eliminating the need for manual tuning while maintaining optimal performance.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system incorporates feedback mechanisms where the autotuning controller continuously monitors circuit states and performance metrics, compares them against target values, and adjusts control parameters accordingly. This closed-loop feedback enables automated optimization, achieving manufacturing precision through iterative adjustment while dramatically improving productivity by eliminating manual tuning processes.

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If digital DPA with autotuning controller is implemented, then adaptability is improved, but device complexity increases

Engineering Contradiction:
Improveadaptability to circuit statesVSAvoidcompensation circuit complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The autotuning controller serves multiple functions within a single integrated unit: it measures circuit states, evaluates performance metrics, calculates optimal control parameters, and adjusts amplifier operations. By consolidating these functions into one universal controller, the system achieves high adaptability while managing overall device complexity through functional integration rather than proliferation of separate components.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Adaptability or versatility

If fixed phase alignment and power splitting ratio are used, then device complexity is reduced, but adaptability to different signal standards deteriorates

Engineering Contradiction:
Improveoperational bandwidthVSAvoidtunable parameter complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent transforms fixed phase alignment and power splitting ratio into dynamic,可调 parameters controlled by the autotuning system. The controller continuously adjusts these parameters based on input signal characteristics and circuit state, enabling the DPA to operate efficiently across different signal standards and bandwidths without requiring complex manual reconfiguration for each standard.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4154398B1Interactive online adaptation for digital pre-distortion and power amplifier system auto-tuning
Publication Date: 2024.08.07 MITSUBISHI ELECTRIC CORP
  • EP4154398B1 patent drawingFigure 1
  • EP4154398B1 patent drawingFigure 2
  • EP4154398B1 patent drawingFigure 3A

AI summary

An autotuning controller is provided for improving power efficiency and linearity of digital power amplifiers (DPAs). The controller includes an interface including input and output terminals connected to the DPAs, the interface being configured to acquire input signals and output signals, a digital pre-distortion (DPD)-DPA adaptive controller including a processor and a memory running and storing a DPD algorithm, an efficiency enhancement method and a learning cost function. The DPD adaptive controller is configured to perform steps of computing DPD coefficients to define a learning cost function based on a DPD model by use of a data-driven optimization method, wherein the leaning cost function includes both variables of a DDA performance and a DPD performance, updating the learning cost function based on the DPD performance, optimizing the updated learning cost function by solving the updated learning cost function with respect to the variables of the DDA performance, and providing optimal parameters for DPA and DPD via the interface.