Wideband envelope tracking power amplifiers with reduced bandwidth power supply waveforms and adaptive digital predistortion techniques

Jinseong Jeong, Donald F. Kimball, Myoungbo Kwak, Chin Hsia, Paul Draxler, Peter M. Asbeck

Research output: Contribution to journalArticlepeer-review

88 Scopus citations

Abstract

This paper presents a new technique to reduce the bandwidth of the dynamic power supply waveform used in wideband envelope tracking power amplifiers (PAs). When the envelope tracking technique is applied to broadband signals such as WCDMA and 3GPP LTE, the wide bandwidth of the envelope signal makes it difficult to implement the dynamic supply modulator efficiently and accurately. We show here a technique to reduce the bandwidth of the power supply waveform, thereby allowing better efficiency for the supply modulator; and a linearization method for correcting the nonlinearity caused by the bandwidth reduction. The feasibility of this technique is demonstrated for a single carrier WCDMA signal with a 7.6-dB peak-to-average power ratio using a GaAs high-voltage HBT PA. The bandwidth of the power supply waveform is reduced from 20 to 4 MHz. After linearization, the reduced bandwidth envelope tracking PA exhibits an average output power of 28W, an average gain of 12 dB and an overall power-added efficiency of 49%. The measured normalized rms error is as low as 0.67% with an adjacent channel leakage ratio of-53.9 and-54.2 dBc at offset frequencies of 5 and 10 MHz, respectively.

Original languageEnglish
Article number5313822
Pages (from-to)3307-3314
Number of pages8
JournalIEEE Transactions on Microwave Theory and Techniques
Volume57
Issue number12
DOIs
StatePublished - Dec 2009

Keywords

  • Adaptive digital predistortion
  • Bandwidth reduction
  • Dynamic supply modulator
  • Envelope tracking
  • Lookup table (LUT)
  • Memory effects
  • Peak-to-average power ratio
  • Power amplifiers (PAs)

Fingerprint

Dive into the research topics of 'Wideband envelope tracking power amplifiers with reduced bandwidth power supply waveforms and adaptive digital predistortion techniques'. Together they form a unique fingerprint.

Cite this