AI Digital Predistortion for Multi-Impairment Transmitter Linearization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predistortion architectures are inefficient in linearizing complex communication systems, such as MIMO and mMIMO, due to the challenges of developing a predistortion model that operates efficiently across multiple input conditions and varying operating conditions.
Innovation Solution
An AI-driven linearization method and system that uses a single set of DPD coefficients across an entire signal operating range, adapting to different and changing operating conditions without the need for continual feedback signals or dynamic re-computation of coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional DPD architectures are used in MIMO and mMIMO systems, then the system can operate with multiple antennas and beamforming, but the linearization efficiency deteriorates due to the complexity of developing predistortion models that operate efficiently across multiple input conditions and varying operating conditions
Solution Approach 1:
The patent applies universality by developing a single predistortion model that functions across multiple input conditions and varying operating conditions in MIMO and mMIMO systems. The model is designed to handle different antenna configurations, beamforming scenarios, and operating parameters simultaneously, eliminating the need for separate models for each condition and thereby improving linearization efficiency while maintaining adaptability.
2Adaptability or versatility
If multiple sets of DPD coefficients are used to cover different operating conditions, then the system can adapt to varying conditions, but the device complexity increases
Solution Approach 1:
The patent merges multiple operating conditions into a single unified predistortion model. Instead of maintaining separate DPD coefficient sets for different operating conditions, the model integrates all conditions into one structure that automatically adapts to the current operating state, thereby reducing device complexity while maintaining comprehensive coverage of varying conditions.
3Measurement precision
If feedback signals are used for continual adaptation of DPD coefficients, then the system can maintain accuracy under changing conditions, but the power consumption and computational burden increase
Solution Approach 1:
The patent applies preliminary action by pre-configuring the predistortion model to inherently account for varying operating conditions. The model is designed during the design phase to automatically adapt to different conditions without requiring continual feedback-driven recalibration, thereby maintaining linearization accuracy while significantly reducing power consumption and computational burden during operation.
Data Source
AI summary
An artificial intelligence (AI) driven linearizer for a transmitter, comprising an input interface for inputting linearizer signals comprising information carrying signals, and operating conditions parameter signals, other than the information carrying signal, wherein the operating conditions parameter signals represent metrics affecting transfer characteristics of the transmitter, over a selected operating range of the transmitter, and a predistortion actuator circuit configured with an AI predistortion model for predistorting at least part of the information carrying signal to produce predistorted signals, the predistortion model being configured to be operable for adaptation to said characteristics of the transmitter using a single set of model coefficients that are unchanged over said entirety of said selected operating range.