pulsed phase noise measurements · 2018. 4. 15. · pulsed phase noise measurements yhp t&m 90...

Pulsed Phase NoiseMeasurements

Ray FriedYHP/Spokane

.:':-'::'j"::'=':::-:'-.'.:~-'=":::.-:.

0000000000

1 0000 :0000 ~'. . .... ,--~_ ..~.. BBBBg:

OD=oClJDD00000000 DO'

RF & MicrowaveMeasurementSymposiumandExhibition

rl/fla HEWLETT~~ PACKARD

www.HPARCHIVE.com

Pulsed Phase Noise MeasurementsYHP T&M 90Tokyo/Osaka

•

Welcome to the

Pulsed Carrier

Phase NoiseMeasurements

Ray Fried YHP

T&M 90Tokyo/Osaka

• This paper discusses Pulsed Phase NoiseMeasurements. Many of the concepts andsolutions presented in this paper weredeveloped and compiled as a result ofhelping customers make these demandingmeasurements on state of the art radarsystems.

Pulsed noise measurements have traditionallybeen made with custom test sets. With theintroduction of the HP 3048, a generalpurpose measurement tool now exists.

• •There is an HP Application Note that coversthis topic in more detail. It is ApplicationNote 386. You may find it as usefulreference material if you have furtherinterest in this topic. If you have Questionsafter the paper is given, there will be timefor Questions. We would also like to inviteyou to the Phase Noise Measurement boothspace for a demonstration of the hardwareafter this presentation.

Pulsed Carrier Phase Noise Measurement This is the agenda we will be following.

Agenda

• BasIcs of pulsed carriers

• Umltatlons with makIng pulsed measurements

• Recommended hardware configurations

• 3048A and 11729C capabilities

• Summary

*

**

***

First a few comments about applicationsof pulsed RF carriersSome brief basicsLimitations with making· pulsedmeasurementsRecommended hard ware configurationsThe HP3048 and HP 11729C capabilitiesThen we will conclude with a summary

1.

www.HPARCHIVE.com

t--------i~ ..._<--

-_.~

~OOt1·,ooc~........

Stepped-Frequency Radar

(Imaging Applications)

Pulsed Carrier Phase Noise Measurement

Agenda

• Applications

• Umitations with making pulsed measurements

• Recommended hardware configurallons


• Summary

Many radar systems use pulsed technology.

Types such as Range Radars, Coherent,Frequency Agile, Chirp, and Phase Coded toname a few.

These may be ground based or airborneradars.

They may be used for many differentapplica tions.

And they cover a wide range of frequencies.

Here is a block diagram of a Radar used forimaging applications. It, like many otherradars, uses pulsed technology.

The quality of a pulsed radar systems isoften directly related to the noise of thepulsed L.C. In this case, you can see theL.C. chain here. The L.C. phase noise canaffect the radars sensitivity & resolution.

Since signal processing components such asamplifiers and pulse modulators can addmultiplicative and additive noise to the CWoscillator, it is necessary to measure the noiseof the pulsed signal.

Now lets talk about some of the basics ofpulsed carriers.

2.

www.HPARCHIVE.com

Basic Characteristics of a Pulsed Waveform

FiTiEl-- --- ------E I

AVGt

TIME-

I will use this slide to establish some termswe will be using. Here we have a repetitiverectangular wave shape.

We will use T to denote the period of therepetitive rectangular waveform and T todenote the pulse duration.

Ml..:CJ:;pok~neV--:,.,cn"'-~~

Periodic rectangular pulse train

,I

Addition of a Fundamental Cosine Wave and

its Harmonics to Form Rectangular Pulses

IA

T1ME-

This slide should remind you of theharmonic content of a periodic rectangularwaveform.

As you know, any repetitive wave shape canbe represented by a fundamental signal offrequency ljT, its harmonics and possiblya DC term. Of course, relative phase andmagnitude and harmonic number areimportant parameters as one attempts todescribe a repetitive waveform by it'sharmonics .

•A\,;O:;:p.lIoo..Ir.,: v,...':.>OON_~(.ll

Af-

.."0

.~a.E..:

"IL:GI~ ...noel..)M.~I¥~,.c.o...

1f

sin xy=

x

2 •f T rFREQUENCY. r

3.

www.HPARCHIVE.com

If we observe a repetitive rectangular signalin the frequency domain we would notespectral lines spaced in frequency atharmonics of IjT. The amplitude of thesuccessive harmonics follow the SinXjXfunction which is the Fourier transform ofthis rectangular function. The envelope ofthe harmonics go through zero magnitudepoints at frequencies that are multiples ofthe reciprocal of the pulse width.

Of course we are assuming a perfectrectangular waveform, that is, one that haszero rise and fall time and no ringing.

Multiplication of a CW Carrier by a pulsed

Waveform Results in a Pulsed Carrier

In reality the pulsed carrier output isgenerated by multiplying a CW carrier signalby a rectangular base band signal.

PulsedWaveform

XCW Carrier

It is important to note the pulse to pulsephase continuity resulting from thisconfiguration. This phase continuity isimportant.

Pulsed Carrier

M(;CJ~,...:Oo...~NIONi.JAl>9CtIGAL

If it were not for this continuity, therewould be large pulse to pulse phase steps orphase jitter. This jitter would likely maskthe noise spectra we are attempting tomeasure.

Also from a baseband noise measurementpoint of view, we will find it necessary tomaintain phase quadrature in the phase noisemeasuring system. This steady state phasequadrature condition would not exist if therewas not pulse to pulse phase continuity.

Carrier FrequencyPRF (Pulse Repetition Frequency)Duty cycle (ratio of T to theInterpulse period) expressed in %

d) Peak powere) Average power =peak power X duty

cycle

At this point we must introduce additionalterms,

a)b)c)

Basic Characteristics of a Pulsed Carrier

Tlme -

MCG/Spokane Dlvlalon,......,.,..,..,... IT.'I HEWLETTI.:ZAI PACKARD

4.

www.HPARCHIVE.com

v ,

,,,I

Convolved with

Now lets look in the frequency domain atthis pulsed carrier signal.

Pulsed Carrier Spectrum

I

Pulsed Wavef~;m Speclru~

v,I,,

,III

I" '.CW Carrier Spectrum

Note the Pulsed Carrier spectrum is that ofthe base band wave shape convolved withthe CW carrier spectrum.

Also note the PRF spaced components andthe signal null at liT.

Lets now focus our attention on how pulsemodulation affects the phase noise of thecarrier.

Mt.CJ:-Po*"'''· L......~I"'/t.IW~IOGAl

When the CW carrier is pulse modulated,each spectral component carries theinformation or noise modulation that waspresent in the original CW frequency. Thisthen effectively lays the spectrum over itselfwith a frequency offset which effectivelyadds noise around the carrier. In effect,noise at any offset from the original CWcarrier will re appear within +/- PRF/2 ofthe original CW signal.

Note that these noise components will bereduced in amplitude by the SinX/Xrelationship of the pulse modulating signal.This added noise component will besignificant if the spectrum of the CW signaldoes not roll off with increasing offset andif the pulse width is relatively short (nulloccurs at large offsets). If the pulse widthis short, there will be many spectral linesthat add to the fundamental component.

\

Combined Noise Spectrum

Carrier with Phase Noise Pedestal

\\

Convolution of Signals

B.i1seband Pulsed Signal

Individual Pulsed Noise Spectra

5.

www.HPARCHIVE.com


Agenda

• Applications

• Basics ot pulsed carriers



• Summary

Now lets look at some of the measurementlimitations when making pulsed noisemeasurements.

6.

www.HPARCHIVE.com

Using a Double Balanced Mixer as a

Phase Detector This is a very common phase noisemeasurement configuration. It uses a phasedetector to convert phase fluctuations tovoltage fluctuations. These voltagefluctuations are then displayed on abaseband spectrum analyzer. Typically, adiode balanced mixer is used as the phasedetector as it has low residuals, rejectsamplitude modulation components and workswith relatively high power. Furthermorewideband units are readily available.

MCCJSPO.....ne v..... ~'onfV_J1'J>"K1I)~

However the use of a mixer for a phasedetector does require the filtering off of themixer input frequencies and sum frequencyat the output. Therefore a LPF is used.Because baseband analyzers do not have lownoise figure front ends, a LNA is typicallyused.

One necessary condition of this approach isthat the mixer must be held in phasequadrature. This can be accomplished byfeedback to one of the sources (A or B). Aphase quadrature condition results in amaximum phase detector coefficient, andminimum amplitude detection coefficient.

In reality, if we were to feed the mixer(phase detector) output to one of the Yeas,we would have a phase locked loop using oneof the sources as the YeO, and the mixer asa phase detector. This locked loop wouldthen maintain phase quadrature even thoughthere is a tendency for the sources to driftin frequency.

In this measurement setup, the systemsensitivity is determined by the "phasedetector coefficient" and the LNA noisefigure. A high phase detector sensitivityresults in greater system sensitivity andimproved dynamic range.

7.

www.HPARCHIVE.com

PHASE DETECTOR OUTPUT FOR A

CW SIGNAL

+v

V'F

(L:,.{>lO _ RF (0. K9)

The phase detector output voltage looks likethis after low pass filtering. The detectoroutput voltage magnitude is a function of theinstantaneous delta phase and the phasedetector coefficient. The phase detectoroutput is a continuous signal. In a pulsedcarrier measurement however, things aredifferent.

Volts -v

MCG/Spokane Dlvl.lonRFIMWJAP9OI14.GAl

rh~HEWlETT~l.I PACKARD

:>hase Detector Output for a Pulsed Carrier+v

* Ph..e Detector Output I. Sampled Version. of CW Ph..e Nol.e* Effective Sen.ltlvity of Phe.e Detector

Scaled by Duly Cycle

* Nol.e abova P~F AII..ed below P:F

* PRF Filter may b. Requlr.d

In this case, the detector output is "sampled".It is not continuous. The output is presentonly over a portion of the time, that is onlywhen the carrier is present.

MCG/Spok.ne D1vl.lonRFIMWJ.<Pr.JlS.GAl

~HEWLETT1o.:'l'..A PACKARD

8.

www.HPARCHIVE.com

Effects of Measuring Noise On Pulsed Carriers

® Reduced Measurment Dynamic Range

® Limited Measurment Offset Range

* Folding of Noise at Large Offsets to the Carrier

* PRF Feedthrough

* Mixer DC Offset - Lack of Quadrature

* Limited Tuning Range

YHP\T&M9O\RF_,

9.

www.HPARCHIVE.com

This produces a number of effects. We willdiscuss several of them.

a) Reduced measurement dynamicrange. This will result because thephase detector sensitivity iseffectively scaled, or reduced by theduty cycle. This reduction in phasedetector sensitivity directly reducesthe measuremen t sensitivity therebyincreases the measurement noisefloor and ultimately reduces themeasurement dynamic range. A 10% duty cycle increases the noisefloor by 20 dB which in turnreduces the dynamic range by 20 dBas compared to a CW measurement.

Another cause of dynamic rangereduction is self demodulation of theL.O. This can be a significantproblem at small duty cycles sincethe desired signal is present for onlya small portion of the time and theL.O. can demodulate itself duringthe entire period thus becoming amajor factor in noise contribution.

b) The second effect of this sampledmeasurement is described bysampling theory. If a band limitedsignal is sampled with equal spacedtime samples, it can be recovered bylow-pass filtering if the fundamentalfrequency of the sampling processis greater than twice the highestfrequency present in the bandlimited signal. Since the phase noiseof a CW carrier is not band-limited,it is being undersampled. Thereforemeasurements made at largefrequency offsets are redundant withlower offsets.

Measurement Offset Range

,.,

,.,

"'/\ /\ ! 1\ /\ /\'- -I -, .

"""· ...1

The limitation of measurement offset rangehas been graphically illustrated in this slide.Notice a signal X(w) with its sidebands.Also notice the sample rate. Notice that ifwe attempt to make measurements at offsetsgreater than (w s)/2 we simply are measuringthe same noise as we did slightly below(w s)/2. In other words we effectively cannotlearn more about the noise profile bymeasuring at offsets greater than PRF/2.


* Reduced Measurment DynamIc Range

* LImited Measurment Offset Range

; ® Folding of NoIse at Large Offsets to the Carrier

® PRF Feedthrough

® Mixer DC Offset - Lack of Quadrature

* LImited Tuning Range

Noise at larger offsets is folded near thecarrier - into the measurement range. As wementioned before, the contribution of thiseffect depends on the duty cycle and thenoise spectra. In general we can estimate thenoise will be degraded by 10 log base 10 ofthe number of spectral lines within the firstnull.

PRF feedthrough will be present. This canoverload the LNA which would result inuseless measurements. Tn any case, PRFenergy will be apparent on the measurementoutput plot.

10.

www.HPARCHIVE.com

Mixer DC Offset may be a problem.

Theoretically when signals of identicalfrequency and a constant phase offset of 90degrees are applied to a mixer, the mixerwill produce a zero volt output. In practice,real mixers often exhibit a dc offset at thisphase quadrature point. This can result fromreflections and insufficient mixer balance.

The dc offset is the deviation from 0 voltswhen the RF and L.a. inputs are atquadrature.

This mixer DC offset can generate severalproblems. It can overload the LNA. On theother hand, frequently we adjust the relativephase of the mixer input signals to force themixer output to a zero volt condition. Thisis often done as an indication of quadratureand to reduce the LNA input bias voltage.This can be a problem however when the DCzero output condition does not correspond toa quadrature condition. Three problems are:

a) one, the phase detector coefficientis reduced. This reduces sensitivity,increases noise the floor, andultimately degrades the dynamicrange of the measurement system.As you can see, the phase slope ismaximum at the true quadraturecondition.

Pulse off

/

\Pulse on

Mixer output

....." 0 f-+-f--H--/-

Mixer DC Offset

MCCtSPOk.)M 0,...,.:"",Al'IO.NI,JJ.P?OIII.COAl

b) And the second problem is increasedamplitude demodulation. Thisoccurs when quadrature is notachieved.

c) Increased PRF feed through.

11.

www.HPARCHIVE.com


* Reduced Measurment Dynamic Range

* Limited Measurment Offset Range

. * Folding of Noise at Large Offsets to the Carrier

* PRF Feedthrough

* Mixer DC Offset - Lack of Quadrature

@ Limited Tuning Range

Tuning Range

vs.PRF (Absolute Measurements)

Peak Tuning Range < PRF/2

The final problem that I would like to talkabout is limited tuning range. Two factorsmust be considered:

a) The phase detector coefficient isreduced by the pulse duty cycle.This reduces the loop bandwidth ofthe control loop that holds the mixerin a phase quadrature condition. Tocompensate for this, the PeakTuning Range (PTR) can beadjusted by inserting gain in thefeedback loop.

b) The control loop bandwidth must besmaller than the PRF rate in orderto maintain stability and a linearcondition.

The resulting effect is that the Peak TuningRange must be less than the PRF/2.

12.

www.HPARCHIVE.com

MCGJSOOluint [)M:1OI'lfISIVWJI'l""»IIGAt.

This slide shows a PARTIAL SOLUTION tosome of the previously mentioned problems.This is an "absolute" noise measurementconfiguration. We use the term "absolute" toimply a noise measurement on a one portdevice. We refer to two port noisemeasurements as residual, in other words,the noise of the input signal is canceled andnot measured in that case.

Here we have a pulsed Device Under Test.Notice in this one port absolute measurementthe pulse modulator on the L.O. input.It is pulsed on at the same time as the DUT.This has several benefits.

a) By turning off the L.O. when theinput pulse is not present, thegeneration of a DC offset voltageduring the off time will not takeplace. Therefore a more accuratephase quadrature condition ismaintained.

b) Furthermore the amount of PRFfeed through will be reduced.

c) The self demodulation of the L.O.AM noise will be eliminated duringthe period when the incoming signal(and L.O.) is pulsed off. This willreduce one component of the systemresidual noise.

(6 (jl LO-RFII)) (K X ,. )<oew T

Pulsed Absolute Measurements

Pul..d OUT

Pulsed Residual Measurements

v"

(Mjl LO-RFII I) (K<oew x ~

The concept of pulse modulating the L.O. isalso applicable to Pulsed ResidualMeasurements. Notice the 2 port DUT andthe common signal source and the pulsemodulators. Again the benefit of pulsingboth ports is reduced L.O. noisecontribution, reduced PRF feed throughproblems and a smaller DC offset situation.

By using these techniques it is possible tomake absolute measurements down to dutycycles of 5 % and residual measurements to1 % duty cycle.

13.

www.HPARCHIVE.com

•

HP3048A 10~?·11t156

·.~.

E9 flF

tf.,1,.

~&u.-,.. .....Jh:..

HPe665AE9H~}..D

MeG/Spokane DivisionT!M9OtRF_2

14.

wwwHPARCHIVE.com

Lets examine PRF feed through for amoment. As was mentioned before,excessive PRF transients or feedthrough willoverload the LNA. This feed through maybe the result of sampling in the presence ofexcessive mixer DC offset. Or it may be theresult of actual phase transients occurring oneither the measured signal or the L.a. nearthe times of amplitude transitions. PRFfeed through can also result from insufficientport-to-port mixer isolation.

This HP3048 block diagram will be helpfulin the PRF feed through discussion. Noticethe inputs, mixer, low pass filter, and phaselocking circuitry. Also notice the AUXMonitor output. This, along with awideband oscilloscope can be used to observethe phase detector output and determine ifthe LNA is overloaded. If the LNA isoverloaded, it can be bypassed with internalswitches.

The HP 3048 phase noise measurementsystem also includes a phase transientdetector at the LNA output. It is used toautomatically bypass the LNA ifpredetermined PRF levels are detected.When the LNA is bypassed, systemsensitivity and dynamic range is reduced.

Since the "AUX Monitor" port follows theLNA one can also determine if the LNA issaturating. When making measurements inthe PULSED MODE, it is possible tooverride the effect of the PRF detector andmanually switch the LNA "IN" or "OUT". Ifthe saturation is not occurring during asignificant portion of the ON period it maybe advisable to .use the LNA.

PRF Feedthrough

"I

1-1I

_ 0.2 RAD

I V\0- r

Pulse Width =5 em -

0-t-----7 0.5 (- K i = 200 mVjRAD _em Horiz =20 mV jem

I I I 1-

This slide shows the maximum limit of phasetransients for acceptable measurements. Ifphase transients exceed .2 radians for morethan 10% of the pulse, the measurementaccuracy may be degraded.

MCGlSpok..ot ()N,SoIO"

troSfUWJA/>"¥Y:Oc..o.L

PRF.F,iltering A PRF filter can be used to reduce PRFfeed through. In this case it is placedbetween the mixer output and the LNAinput. If a filter is required, chose one thathas a 3 dB cutoff frequency approximatelyequal to PRF/2 and has at least 60 dB ofattenuation at 2 times the PRF.

•

MC(JSookaot ()r.t~0t'I

~tvW~~"'l

HP3048A jO~?·11t15o

AVX ~'::'l!ltl

••.t.

.....HP8665AEY--F---.J

HI"%)

MCG/Spokane DivisionT&M9OIRF_2

,,,H'I!I1J

15.

www.HPARCHIVE.com

When a PRF filter is required, an externalphase detector must be used since there is nodirect access to the output of the internalphase detector.

Notice the "Noise Input" port. It is use whenan external mixer and PRF filter arerequired .

Experience has shown that for many pulsedmeasurements, the measurement can be madewithout a PRF filter.


Agenda

Now lets look at the recommended hardwareconfigurations for a number ofmeasurements.

• Applications

• Basics of pulsed carrIers

• Umilations with making pulsed measurements


• Summary

Recommended Configuration for Absolute

Measurements If PRF Filter is Required

For absolute measurements that require aPRF filter, this configuration isrecommended.

L------------------lTurMIVotll~

HP 304aA

HolM Input

Erl.,nal Ph....Oeleoctor .nd

PRFAn.,- --,III

PUIMdOt/T·

MC~~MO""~

rtl'fWW~"G"'l

• Recommended Configuration. for Residual

Measurements If PRF Filter is RequiredFor residual measurements that require aPRF filter, use this configuration.

HP 30484

Ext.rnal Ph•••D.tKtot and

PRFAtt...- ---,III

HolM Input

16.

www.HPARCHIVE.com

Recommended Configuration for Residual

Measurements If no PRF filter is required, use thisconfiguration for residual measurements.

MCGlS~aneOM"Ol"l~_JitP9<n(.C-tt.


Agenda

• Applications

• Basics of pulsed carriers

• Limitations with making pulsed measurements


• Summary

Now we will look at the HP3048A andHP 11729C capabilities.

• Noise Floor Using an HP 8663A as the

Source

HP :J048A

The measurement system noise floor can beevaluated with this measurement set up.

17.

www.HPARCHIVE.com

ReSIO, NOIse F"LOOq, 86GJR, PRr Z0KHZ. 18% DC0 ... 1 lB. 'A CHr''''1 1.'(·'.... 14 II<" IS" "11I:le - Oll)lee

Residual Noise FloorNotice the -140 dBC floor, even with a 10%duty cycle. The PRF is 20kHz. Usefulmeasurements can only be made to offsets< 10kHz.

'0\'",.

-I. f------+-----t------+-----1-,. f------+-----t-----+-----1-J. f------+-----t-----+-----1-•• f-------f-------I------+-----J-,.F-------+-----f-----+---~-rl.,.F-------+-----f-----+-..--+-+-+l-l. f------+-----f-----+-ll--+-l--+l-I. f------+-----f-----+-ll--!-l-+l.,. f------+-----t-----+-f+--!-l-+l

-181' f-------!-------I------+-H----Jl-l~

~1I8 f------+-----Jf--------j.--AJ-I---Il--f--H·ue f----+---h-,-,----If--------j.--Hf+---Il--+-H

-"'E~~2~~-1'''1 11\ II ~ ,.J I

-". .~·1'" f------+-----If--------j.-----.--I.". L...~-~~-:-!,,~.-----cJIK':----~-:-!,•..,."--~~...J

! «() (dBc ....Hz) ..... ((Hz)

•

MCGlSpok.rtC ()..,.,soO"p;~t.(O.I.L

Residual Noise of, _~P 8447D AmplifierThis configuration is being used to measurethe pulsed residual noise of an HP 8447Damplifier.

Residual Noise of HP 8447D Amplifier•

10100 II< 1011::

Leo' Cd8c:/1o"h:J vs fC,",,%) 100<

Here are the results of the measurement.

18.

www.HPARCHIVE.com

This configuration was used to measure thenoise quality of a pulsed HP 8663A signalgenerator.

This is the recommended configuration forabsolute measurements of pulsed sources.

.lUX Monl\o(Trlg04'f

HP 8663A versus HP 8663A Pulsed

. Source Measurement HP)O......

'-===..:.:.==..::.:.1:"'-----------1 Source Output 10 MHzA

•

III

HP 8663A versus HP 8663A Pulsed

Source Measurement(~" H.'A C.""r1 1.(-' .. t ".,. un 11,31,2' - Ut43,f,l

It

,.. IK lilt 1M< I"

!.«() (dltc ....Hz) vs flHI]

II. ....,

Here are the results of the measurement.Notice the noise floor at 1 to 10Khz offsetis about 10 dB higher than in a non pulsedapplication. For offsets < 100 Hz theperformance is about the same as non-pulsedmeasurements.

• Absolute Measurements with the

HP 3048/11729CHP 304....

IFlMpvt~_~=..:;5-12'0 MHz

8-40 MHzl"

HP 1"U/3Awith Opl D03

640 MHz Out RF0u1~-+=--...J

The HP11729C is a low noise downconverter that can be used for makingabsolute measurements. It uses a 640 MHzreference to convert incoming microwavesdown to a frequency range of 5 to 1280MHz. The 8662 is used for the L.O.

19.

www.HPARCHIVE.com


Agenda

• Applications

• Basics of pulsed carriers

• Limitations with making pulsed measurements



• !:t.IINJ{fiulry

In summary then we have covered some ofthe basics and several limitations in makingpulsed noise measurements. We presentedseveral solutions to these limitations andmade some hardware recommendations.Finally the HP3048 and HP 11729Ccapabilities were presented

Limitations

• Measurement Offset Range

• Mixer DC Ollset

• LO AM Noise

• PRF Feedthrough

• Duty Cycle

Here are some of the limitations wediscussed.

• Summary ContinuedCapabiflliu

HI' JO.t'A: Absolute Ruldtnl

fttJ:Ut.ncyunoe Determined by phUl deteC10t (lnluN! 01 hterfUl)

loIinim.m duly eyel. I" I I"~PRF 2 X PTR Ho Minimom

NO.. FIoo< Phue deltctOf noise suled by duty cycle

/=sIC'( ~2cle .01 HI 10 1MHz olfsets~ t dB 1MHllo.cO MHz ottscu

H?JO<U/1172<J:

frtq.)Cnty r1nQt 10 MHIla 18 GHz. to mm with sPKials MIA

V....wnumdul)'cyd. I" MIA_m2 X PTR MIA

,..;,. fJoo< NP 8662/11 129 CW nO" loId'" by

pU"'- MIA

/=s>cy :: 2 dB .01 Hz to 1MHz onsets MIA::.c dB 1 MHz to .cO MHz: otlsels

I would like to conclude with this slide thatsummarizes capabilities of configurations forabsolute and residual measurements. TheHP3048 is guarantet::d to operate correctly atthese recommended minimums. If ameasurement is made below theseminimums, and the system operatescorrectly, then the accuracy of themeasurement is not in question. You can seethe duty cycle limitations listed here. It maybe possible to make measurements with alower duty cycle if a locked PLL conditioncan be maintained.

20.

www.HPARCHIVE.com

pulsed phase noise measurements · 2018. 4. 15. · pulsed phase noise measurements yhp t&m 90...

Documents