Power Amplifier Predistortion Using Reduced Sampling Rates in the Forward and Feedback Paths.

Area of Science:

• Electrical Engineering
• Wireless Communication Systems
• Signal Processing

Background:

• Next-generation wireless communication demands higher data rates, leading to increased bandwidth utilization (e.g., 400 MHz in 5G NR).
• Existing digital predistortion techniques face challenges with high sampling rate requirements, impacting efficiency and practicality.
• Developing predistortion methods operating at lower sampling rates is crucial for advanced wireless systems.

Purpose of the Study:

• To present a novel predistortion technique that divides processing between digital and analog domains.
• To reduce the sampling rate requirements for both forward and feedback paths in digital predistortion systems.
• To enable efficient and practical predistortion solutions for high-bandwidth wireless communication.

Main Methods:

• A hybrid predistortion approach combining analog and digital domains was developed.
• The analog domain incorporates a memoryless AM/AM gain function.
• The digital domain utilizes a nonlinear model with memory effects to compensate for impairments.
• The sampling rate reduction was applied concurrently to the forward path's DAC and the feedback path's ADC.

Main Results:

• Experimental validation was performed using 20 MHz and 40 MHz 5G signals.
• A 50% sampling rate reduction was achieved for the 20 MHz signal with minimal linearization degradation.
• A 30% sampling rate reduction was achieved for the 40 MHz signal with minimal linearization degradation.
• The proposed technique effectively relaxes sampling rate requirements without significant performance compromise.

Conclusions:

• The novel hybrid predistortion technique successfully reduces sampling rate requirements in wireless communication systems.
• This approach offers a practical solution for implementing efficient digital predistortion in next-generation wireless standards like 5G.
• The method demonstrates a viable trade-off between sampling rate reduction and linearization performance.