Baseband Predistortion Linearization of Power Amplifier - PowerPoint PPT Presentation

1 / 15
About This Presentation
Title:

Baseband Predistortion Linearization of Power Amplifier

Description:

Baseband Predistortion Linearization of Power Amplifier. Lei Ding, G. Tong Zhou ... Better spectral efficiency, need linear power amplifier, low power efficiency ... – PowerPoint PPT presentation

Number of Views:570
Avg rating:3.0/5.0
Slides: 16
Provided by: csip9
Category:

less

Transcript and Presenter's Notes

Title: Baseband Predistortion Linearization of Power Amplifier


1
Baseband Predistortion Linearization of Power
Amplifier
  • Lei Ding, G. Tong Zhou
  • CSIP seminar, April 20

2
Outline
  • Motivation
  • Power Amplifier Modeling
  • Predistortion
  • RF/DSP Testbed
  • Summary

3
Motivation
  • Recent wireless applications prefer linear
    modulation schemes
  • CDMA(IS-95) QPSK
  • OFDM QAM
  • Better spectral efficiency, need linear power
    amplifier, low power efficiency
  • High efficiency nonlinear power amplifier
  • Spectral regrowth in adjacent channels

4
Power Amplifier Modeling
  • Memoryless nonlinear modeling

PA
5
Example
6
Predistortion
PA
Pred
7
Example
8
Example
9
Adaptive Predistortion
  • In real applications, PAs AM/AM and AM/PM are
    not available and are changing with time.
  • Adaptive predistortion is needed.

10
RF/DSP Testbed
11
Table-based adaptive pred.
Given a block of baseband samples, rn rn
1-(k-n)CrEm/Nr ?n ?n
1-(k-n)C?Ep/Nr rn1 rn1 (k-n)CrEm/Nr
?n1 ?n1 (k-n)C?Ep/Nr Where Cr and
C? are constants to control convergence rate and
stability, Nr is the total number of samples
falling into the same amplitude interval rn
rn1.
12
Results
13
Results
14
Results
15
Summary
  • We developed a flexible RF/DSP testbed, which
    allows us to test a variety of algorithms and RF
    devices.
  • We implemented a table-based predistortion
    linearization algorithm and achieved around 9 dB
    suppression on spectral regrowth.
  • Next, we will carry out real-time
    implementation and include memory effects in the
    predistortion algorithms
Write a Comment
User Comments (0)
About PowerShow.com