1 MATLAB Digest www.mathworks.com

By Sean Lynch, Nujira

Power amplifiers (PAs) ensure that the source radio frequency (RF) signal,

such as a DVB, 3G, LTE, or 4G signal, is powerful enough for transmission.

Because PAs use a fixed supply voltage, they draw maximum power whatever

the amplitude of the signal, making them notoriously inefficient for amplitude

modulated RF signals.

Verification of High-Efficiency Power Amplifier Performance at Nujira

MATLAB Digest

Products Used ■ MATLAB®

■ Simulink®

■ Communications Toolbox™

■ Control System Toolbox™

■ Fixed-Point Toolbox™

■ Instrument Control Toolbox™

■ MATLAB Compiler™

■ Signal Processing Toolbox™

Figure 1. Left: Power amplifiers with a fixed supply voltage. Right: Supply voltage modulated using envelope tracking.

Conventional PAs waste as much as 80% of the energy they consume as

dissipated heat. The PAs in a cellular base station, for example, account

for half the total power consumed.

Although it was first described more than 60 years ago, envelope tracking

has not been applied commercially until recently, largely due to the diffi-

culty of implementing a power supply modulator that meets the efficiency,

bandwidth, and noise requirements of wideband signals such as multicar-

rier WCDMA, WiMAX, or DVB.

Nujira’s envelope tracking technology can double the efficiency of PAs

and dramatically reduce power dissipation, which lowers energy bills and

substantially reduces the amount of cooling required (Figure 1). It also

enables higher device output power, allowing broadcasters to extend the

range of existing broadcasting towers. Lastly, the wide-band operation

of Nujira’s technology enables broadcasters to use fewer PA designs to

cover their target broadcast spectrum.

The Benefits of Envelope Tracking Technology

Envelope-trackingtechnologycandoubleaPA’sefficiencyfrom30%to60%ormorebydynamicallymodulating theamplifier’s supplyvoltageaccording to theRFsignalpassingthroughthedevice.

Nujira’sHighAccuracyTracking(HAT™)technology,thefirstpracticalimplementationof envelope tracking, isbeingusedorevaluated forcellular infrastructureand terminals,broadcast transmitters, military communications and other applications. It is featured inthreeproductlines:Coolteq.Lformobilehandsets,Coolteq.hforcellularbasestations,andCoolteq.uforDVBtransmitters.WeusedMathWorkstoolsacrosstheentiredevelopmentlifecycleoftheseproductlines,fromresearch,throughmodulatordesign,toverificationandautomatedtesting.

2 MATLAB Digest www.mathworks.com

Nujira Power Modulator Technology Toimplementenvelopetracking,amodulatormustkeeptheamplifieroperatingatitsmostefficientpointbychanging thePA’ssupplyvoltageinphasewiththetransmittedsignal.Forabasestationtransmittingamodulatedsignalwitha40megahertz(MHz)envelope,for example, the supply voltage of the PAmustbechangedatafrequencyofatleast40MHz,anditmustbepreciselyalignedwiththetransmittedsignal.

In the product design phase, we usedSimulink®, Communication Toolbox™ andControlSystemToolbox™tomodelthekeypower modulator circuitry and then ransimulations to project the performance ofvarious modulator designs. The modula-tormodelincludedkeycircuits,suchasthe

switchmodepowerstageandlinearcorrec-tionamplifier.Simulating theoperationofthemodulatorfromtherawenvelopeinputto the processed envelope output enabledustopredictthemodulator’sefficiencyandtrackingaccuracy.Wethenusedthesimula-tionresultstoverifytheperformanceofthemanufacturedproduct.

Developing and Validating Digital Predistortion AlgorithmsModulatingthegatevoltageofthePA’stran-sistor in real time distorts the transmittedsignal,aneffectthatcanbecorrectedusingdigital predistortion. Our digital predis-tortion algorithms are applied within thetransmissionchaintoproducealinearizedoutput that meets adjacent channel power(ACP)requirementsforallrelevantstandards.

Wedevelopedandtestedfloating-pointver-sionsofthedigitalpredistortionalgorithmsusing MATLAB® and Signal ProcessingToolbox™andthenconvertedthemtofixedpointusingFixed-PointToolbox™inprepa-ration for implementation on an FPGAandDSP.Wedevelopedasetoftestinputsandexpectedoutputs tovalidate thealgo-rithmsbeforeimplementingtheminhard-ware.Later,wereusedthesametestvectorsto verify that the hardware produced thesameresultsas thefixed-pointMATLABimplementation.

Modulator Design ValidationIn the first phase of validation, we useda test configuration that included a testbaseband,themodulator,andalargeresis-tive load. The validation bench compriseddigital multimeters (DMM), power sup-plies,andanoscilloscope(Figure2).AlltheequipmentwascontrolledusingMATLABandInstrumentControlToolbox™,typicallyvia Standard Commands for Program-mable Instruments (SCPI) over a GPIBbus.Thetestteamcapturedarangeof2G,3G,LTE,andTDMAtestsignalsusinganAgilentMXAsignalanalyzerandthenusedMATLABtoperformcrestfactorreductiononthesignalstopreparethemforuseinthetestenvironment.

UsingMATLABandMATLABCompiler™we built a standalone control application,which we used to drive the tests. This ap-plicationconfiguresthebasebandviaaUSBMEX-fileinterfaceandcontrolsthemodu-latoroveritsI2Cinterface.DuringatestrunweusedInstrumentControlToolboxtocap-ture theoutputwaveforms fromthescopeandthenranMATLABscriptstocomparethe captured waveform against the trans-mitted signal to verify that the modulatoraccuratelytrackedthesignal.

Inthenextroundofvalidation,wereplacedFigure 2. The 2.1GHz RF validation test bench.

3 MATLAB Digest www.mathworks.com

the resistive load with an actual PA andadded a spectrum analyzer and two USBpower meters to the validation bench. Withthis configuration we can test the full RFsetup.Thedevelopmentkitthatweprovideto potential customers contains this sameRFsetup.ItincludestheCoolteqDevelop-ment System and a standalone MATLABapplication with a graphical interface thatenables customers to control every aspectofthetests(Figure3).

System-Level ValidationTheNujira testing teamhasbuilt anauto-mated test equipment (ATE) system forthe complete baseband, modulator, andPAchain.Thetestsetupincludesmultiplepowersupplies,powermeters,andDMMs,inadditiontoanoscilloscopeandspectrumanalyzer(Figure4).Thesetupenablesustoswap out the baseband, modulator, or PAto quickly evaluate a new design. We useMATLABtodrivetheautomatedtestsand

control the test instruments. We capturewaveforms from the scopes and basebandreceiversandthenrunMATLABscriptstoanalyze theperformanceof theentire sys-tem at various operating points, including

arangeofvoltagerails,andattemperaturesbetween-40to+60degreesCelsius.

AtypicaltestrunforoneRFwaveformre-quires about 50 measurements. Becausemanualtestingrequirestheengineertosetupregistersforeachtest,asingleruncouldtake several hours to complete manually.With our ATE, engineers can complete asingleruninseveralminutesandafullre-gression test for 10 waveforms in 45 min-utes.Moreover,theresultsfromtheATEaremuchmoreconsistentandrepeatable thanthemanualtestresults.TheMATLABbasedATEhassavedNujiramorethan£1millioninmanualtestingcoststodate.

The Coolteq.h product line is already inproduction, and we are now testing thefirstCoolteq.Lchipsformobilehandsets.MathWorkstoolshavehelpedNujirabringtheseproductsfromrawresearchtoproduc-tionsystems,andhaveenabledustodevelopacompleteATEsystemforcomponent-andsystem-leveltesting.■

Figure 3. Left: Waveform analysis display generated from the MATLAB based Coolteq Development System. Right: Coolteq Development System control panel.

Figure 4. Nujira’s automated test equipment setup for system-level validation.

4 MATLAB Digest www.mathworks.com

Resources

visit www.mathworks.com

technical support www.mathworks.com/support

online user community www.mathworks.com/matlabcentral

Demos www.mathworks.com/demos

training services www.mathworks.com/training

thirD-party proDucts anD services www.mathworks.com/connections

Worldwide contactswww.mathworks.com/contact

e-mail info@mathworks.com

91821v00 00/10

© 2010 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be trade-marks or registered trademarks of their respective holders.

For More Information ■ TCP/IP Communication for Real-Time

and Embedded Systems www.mathworks.com/tcp-ip

■ Automating the Implementation of Software Defined Radios at Northrop Grumman www.mathworks.com/sdr-northrop

■ Nujirawww.nujira.com