Steven Baker Founder & Director OpenET Alliance RF signal considerations 26 Complexity of amplifier model (behavioral or transistor-level?) Signal Complexity (primarily length) and

Welcome

1

© 2012 Agilent Technologies, Inc.

Andy Howard

Senior Application Specialist

Agilent EEsof EDA

Steven Baker

Founder & Director

OpenET Alliance

Outline

Steven Baker, OpenET Alliance

• What problem are we trying to solve?

• What is Envelope Tracking?

• Envelope Tracking Market Status

Andy Howard, Agilent EEsof EDA

• Simulating Envelope Tracking

• What type of amplifier model?

• Characterizing the power amplifier

• Modulation signals

• Preliminary simulations

• Investigating non-idealities

2


Envelope Tracking—An Introduction


What is the OpenET Alliance?

A non-profit corporation that operates as an open membership industry

association

With a mission to drive the development and adoption of envelope

tracking techniques, components and products in the wireless

communications industry, to deliver more efficient RF transmitters

We do this by explaining the benefits of envelope tracking and enabling

the ecosystem through standardisation, tools, research and networking

Members include baseband & transceiver vendors, PA vendors, power

modulator vendors, OEMs and leading academics

www.open-et.org

Page 4

Introduction

© Copyright 2012 OpenET Alliance

What’s the problem?

What’s envelope tracking and how does it help?

Market status

© Copyright 2012 OpenET Alliance Page 5

Outline

Introduction

Increasing Peak-to-Average Power Ratio (PAPR)


Standard Launched

Spectral Efficiency (bpsHz

-1)

PAPR (dB)

Fixed Supply

Efficiency (typ. %)

GSM 1991 0.17 0 65

WCDMA 2003 0.1 3.4 45

HSUPA 2008 1. 5 6.5 35

LTE 2010 4 8 30

What’s the Problem?

Introduction

Increasing band fragmentation

As PAPR goes up PA efficiency goes down

Band fragmentation broadband operation even lower efficiency


Envelope Tracking at 50,000 Feet

Introduction

Constant supply Variable envelope

Variable supply

Variable envelope


Instantaneous Efficiency

Introduction

60

50

40

30

20

10

0

Devic

e E

ffic

iency (

%)

Pout (dBm) 40 30 20 10 0 -10

1.0 V 1.5 V 2.0 V 2.5 V 3.0 V 3.5 V 4.0 V 4.5 V 5.0V

Fixed supply efficiency locus Envelope Tracking efficiency locus

Signal PDF

Average

Peak

PAPR ~8dB


Power Amplifier Q-DAC

I-DAC

To Rxr

Duplexer I

Q

Vbatt


Introduction


Power Amplifier Q-DAC

I-DAC

To Rxr

Duplexer I

Q

Env- DAC

Magnitude calculator

Pre-envelope

gain

Envelope shaping

Envelope Tracking

Modulator

Vbatt

Post-envelope

gain & offset

Delay

B

A


Introduction

Market Status

LTE Infrastructure

• Successfully deployed in fully commercial 40W remote radio heads

LTE Terminals

• All major silicon platform vendors are working on ET solutions today

• If you’re working on LTE terminals or LTE terminal technologies you’ll be

working with envelope tracking in the next year to 18 months

LTE Small Cells

• Growing interest in ET as a solution for small cells—power consumption,

thermal management, ‘broadbanding’

www.open-et.org

Page 11

Introduction

© Copyright 2012 OpenET Alliance

Outline


• What problem are we trying to solve?

• What is Envelope Tracking?

• Envelope Tracking Market Status

Andy Howard, Agilent EEsof EDA

• Simulating Envelope Tracking

• What type of amplifier model?

• Characterizing the power amplifier

• Modulation signals

• Preliminary simulations

• Investigating non-idealities

12


Simulating Envelope Tracking

13

Sample

Input Signal

Power

Apply Shaping

Load

Modulated

Input

Signal

Modulate

Drain

Voltage

Modulate drain voltage in response to input signal envelope

Shaping curve attempts to maintain constant gain or gain compression


Outline


What type of amplifier model?

Characterizing the power amplifier

Modulation signals

Preliminary simulations

Investigating non-idealities

14



15

Do you have a transistor-level amplifier model?

If you just have a physical amplifier, can you measure its X-parameters?


Outline




Modulation signals



16



17

Checking for memory effects and bandwidth

Obtaining data for constant-gain shaping

Obtaining data for constant-gain-compression shaping


Testing for memory effects

18

Inject two tones; sweep their spacing

Inequality between lower and upper intermodulation distortion tone

amplitudes => memory effects


Testing for memory effects (a different amplifier)

19

Memory effects significant for tones > ~100 kHz spacing

Amplifier may still work well for narrower band signals


An alternative way of testing for memory effects –

using Envelope simulator

20

2 tones,

spaced

1.5 MHz

2 tones,

spaced

150 kHz

Vload phasor

versus time


Obtaining data for constant-gain shaping table

21

Simulate amplifier with HB;

sweep both Pavs (available

source power) and

drain bias

Specify desired gain:

If available source power = 10.5 dBm, set drain bias to 3.5 V to keep gain = 11 dB


Obtain new shaping table for different gain,

instantly

22

If available source power = 10.55 dBm, set drain bias to 3.0 V to keep gain = 10 dB


Obtaining data for constant-gain-compression

shaping table

23

Desired gain

compression

set to 1.5 dB:

Drain bias versus source

power to maintain

constant gain

compression

Available Source Power, dBm

Dra

in B

ias V

olta

ge

Drain Bias

Voltage


Obtain new shaping table for different amount of

gain compression, instantly

24

Desired gain

compression

set to 2.0 dB:

Drain bias versus source

power to maintain

constant gain

compression

Available Source Power, dBm

Dra

in B

ias V

olta

ge

Drain Bias

Voltage


Outline




Modulation signals



25


Modulated RF signal considerations

26

Complexity of

amplifier model

(behavioral or

transistor-level?)

Signal Complexity (primarily length)

and measurement

(PAE, AM-to-AM, AM-to-PM, or

specification-compliant EVM?)

Shorter simulation time

Longer simulation time

Late in design process: use longer signals, obtain specification-compliant results

Early in design process: use short signals, obtain quick results, make changes


Obtaining modulated RF signals

27

Numerous sources exist in Ptolemy Wireless Libraries

I and Q baseband data may be modulated onto RF carrier

Obtain from Agilent Signal Studio software

Example of LTE signal and its statistics


Outline




Modulation signals



28


Short simulation with fixed bias

29


Fixed bias simulation results

30

Power delivered to load, dBm


Short simulation with envelope tracking bias

31

Ideal, behavioral components detect input power,

read corresponding drain bias from shaping table in file


Simulation results with envelope tracking bias

32

Power delivered to load, dBm


Things to consider

33

Power amplifier gate bias and minimum drain bias

Amplitude of input modulation signal

Type of shaping table – constant gain or gain compression

Performance sensitivity to external source and load

impedances


Ptolemy co-simulation with a Vector Signal

Analysis sink

34


Outline




Modulation signals



35


Modeling a finite slew rate in the bias modulator

36

Slew rate =

40 V/microsecond

Slew rate =

10 V/microsecond


Modeling a time delay difference between RF and

bias modulation signals

37

Time delay delta =

0 nanoseconds

Time delay delta =

20 nanoseconds


Summary

38

Agilent ADS – well suited for:

• Modeling power amplifiers

• Investigating power amplifier performance

• Generating and simulating modulated signals

• Investigating envelope tracking schemes

• Modeling various non-idealities


For more information

39

Download these examples from the Agilent EEsof Knowledge Center: http://edocs.soco.agilent.com/display/eesofkcads/Applying+Envelope+Tracking+

to+Improve+Efficiency

Application note on envelope tracking simulation:

http://cp.literature.agilent.com/litweb/pdf/5991-1463EN.pdf

On envelope tracking:

http://www.open-et.org/

Or contact me directly: [email protected]


http://edocs.soco.agilent.com/display/eesofkcads/Applying+Envelope+Tracking+to+Improve+Efficiency











You are invited

Dr. Peter H. Aaen

RF Modeling and Measurement

Technology Team

Freescale Semiconductor

You can find more webcasts

www.agilent.com/find/eesof-innovations-in-eda

www.agilent.com/find/eesof-webcasts-recorded

http://www.agilent.com/find/eesof-innovations-in-eda







http://www.agilent.com/find/eesof-webcasts-recorded





Steven Baker Founder & Director OpenET Alliance RF signal considerations 26 Complexity of amplifier model (behavioral or transistor-level?) Signal Complexity (primarily length) and

Documents