Linear Predistortion-less MIMO Transmitters

Joep Osger Pepijn Zanen

Research output: ThesisPhD Thesis - Research UT, graduation UT

86 Downloads (Pure)

Abstract

The ideal amplifier multiplies an input signal with a desired gain, creating a linear relation between the input and output levels. However, in real-world power amplifiers in transmitters for wireless communication, this relation is never perfectly linear. This distortion degrades signal integrity and causes unwanted radio emissions, interfering with other frequency bands. Common approaches to increase system linearity are to improve the inherent amplifier linearity, typically at the cost of other amplifier performance metrics, or to predistort the input signal with the inverse distortion of the amplifier, such that the two cancel each other. The required calculations for this pre-distortion do however consume power, especially in multi-antenna transmitters at high data rates.

In this dissertation, Linear Predistortion-less MIMO Transmitters, Joep Zanen models the power consumption for digital predistortion, indicating this can exceed that of the power amplifier in multi-antenna transmitters. This motivates the need for high linearity power amplifiers without predistortion. These amplifiers are subsequently implemented on-chip using switched-capacitor power amplifiers with novel constant output conductance drivers, achieving state of the art linearity for their power consumption. Further experiments with added quadrature mismatch calibration circuits prove to remove the need for digital compensation altogether. The performance as a high-datarate multi-antenna transmitter system is demonstrated for applications with directional radio transmission using beamforming, allowing for faster connections and more simultaneously communicating devices in a wireless network, while decreasing transmitter power consumption.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • University of Twente
Supervisors/Advisors
  • Nauta, Bram, Supervisor
  • Klumperink, Eric A.M., Supervisor
Award date23 Nov 2023
Place of PublicationEnschede
Publisher
Print ISBNs978-90-365-5886-0
Electronic ISBNs978-90-365-5887-7
DOIs
Publication statusPublished - 23 Nov 2023

Fingerprint

Dive into the research topics of 'Linear Predistortion-less MIMO Transmitters'. Together they form a unique fingerprint.

Cite this