Application Note 99 May 2005: Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1

Application Note 99
May 2005
LT5528 WCDMA ACPR, AltCPR and
Noise Measurements
Doug Stuetzle
INTRODUCTION
ACPR (adjacent channel power ratio), AltCPR (alternate
channel power ratio), and noise are important performance metrics for digital communication systems that
use, for example, WCDMA (wideband code division multiple access) modulation. ACPR and AltCPR are both
measures of spectral regrowth. The power in the WCDMA
carrier is measured using a 5MHz measurement bandwidth; see Figure 1. In the case of ACPR, the total power
in a 3.84MHz bandwidth centered at 5MHz (the carrier
spacing) away from the center of the outermost carrier is
measured and compared to the carrier power. The result
is expressed in dBc. For AltCPR, the procedure is the
same, except we center the measurement 10MHz away
from the center of the outermost carrier.
To measure ACPR and AltCPR, refer to the test setup

shown in Figure 2. The DUT (device under test) is the
LT5528, which is a high linearity direct I/Q modulator. It
accepts WCDMA modulation at the baseband inputs, and
generates a WCDMA modulated signal at the RF output.
Note that a free running RF generator provides the LO
signal. This type of generator is used because of its
superior noise performance. This is critical, as a noisy LO
signal may corrupt the modulator output, and consequently the ACPR measurement. For the generator shown,
the automatic level control must be switched off to avoid
degrading its broadband noise floor. Also, the operating
frequency can drift slightly, so manual frequency correction could be needed.
, LTC and LT are registered trademarks of Linear Technology Corporation.
All other trademarks are the property of their respective owners.
1 TO 4 CARRIERS
3.84MHz WIDE
SPACED 5MHz APART
3.08MHz TO
6.92MHz:
45dBc/30kHz
8.08MHz TO
11.92MHz:
50dBc/30kHz
FREQUENCY
0
MHz
5
MHz
10
MHz
Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1
AN99-1
Application Note 99
I-DAC
5V POWER
SUPPLY
BALUN BOARD
MINI-CIRCUITS
AD T2-1T
TTE LE1182
VCC EN
ATT
VCC
LT5528
BBIP
TTE LE1182
BBIM
1dB
STEPS
V2I
RF OUT
I-CHANNEL
RF = 1.8GHz
TO 2.2GHz
0
Q-DAC
90
ATT
BBQP
AGILENT
E4436B
W-CDMA
EXTENSION
BBQM
1dB
STEPS
BALUN BOARD
MINI-CIRCUITS
AD T2-1T
SPECTRUM ANALYZER
FSEM30
ROHDE AND SCHWARZ
HPIB
BALUN
Q-CHANNEL
PC
V2I
GND
LOIN LO = 1.8GHz
TO 2.2GHz, 0dBm
HP8616A
SIGNAL GENERATOR
AN99 F02
Figure 2. ACPR Measurement Setup
In general, the baseband source will not be ideal. It

generates spectral regrowth and noise which may swamp
the performance of the DUT. The lowpass filters shown at
the baseband generator outputs reduce these impairments to a tolerable level. Filters suggested for this purpose are listed below:
1-channel ACPR measurement

Filter part number LE1182 2.5M-50-720B, from
TTE Engineering
30
40
50
60
70
80
Rejection at 3.08MHz >80dB
120
2.1275

Filter part number LE1182 10M-50-720B, from
TTE Engineering
AN99-2
NOISE
CORRECTED
100
110
SPECTRUM
ANALYZER
90
Filter part number LE1182 5M-50-720B, from TTE

Engineering
An accurate measurement of the spectral regrowth of a

highly linear device such as the LT5528 is difficult because
its dynamic range may rival that of the measurement
equipment. Because of this, it is important to account for
the noise of the measurement system; i.e., the spectrum
analyzer. Refer to Figure 3.
POWER IN 30kHz BW (dBm)
The spectrum analyzer must have a wide dynamic range.

That means a high input 3rd order intercept point, and a
low noise floor. The analyzer shown in Figure 2 meets both
of these requirements.
NOISE FLOOR
2.1325 2.1375 2.1425 2.147.5 2.1525
RF FREQUENCY (GHz)
AN99 F03
Figure 3. ACPR Spectrum for a Single Carrier WCDMA Signal
To do this, first measure the noise floor of the spectrum

analyzer with a 50 input termination. The input attenuation of the analyzer should be set to 0dB. This will minimize
the input 3rd order intercept point of the measurement
system, as well as the noise figure. A 30kHz resolution
bandwidth is used because the spectrum analyzer shown
has the lowest noise figure (about 24dB) at that resolution
bandwidth. This spectrum analyzer includes an RMS
display detector mode, which is specifically designed to
Application Note 99
I-DAC
5V POWER
SUPPLY
BALUN BOARD
MINI-CIRCUITS
AD T2-1T
TTE LE1182
VCC, EN
ATT
VCC
LT5528
BBIP
BBIM
1dB
STEPS
Q-DAC
RF = 1.8GHz
TO 2.2GHz
0
90
ATT
BBQP
AGILENT
E4436B
W-CDMA
EXTENSION
RF OUT
I-CHANNEL
TTE LE1182
BBQM
1dB
STEPS
BALUN BOARD
MINI-CIRCUITS
AD T2-1T
Q-CHANNEL
BALUN
V2I
GND
SPECTRUM ANALYZER
FSEM30
ROHDE AND SCHWARZ
HPIB
K&L 5BT SERIES

BANDPASS FILTER
SET TO REJECT SIGNAL
PC
LOIN LO = 1.8GHz
TO 2.2GHz, 0dBm
HP8616A
SIGNAL GENERATOR
AN99 F04
Figure 4. 30MHz Offset Noise Measurement Setup
measure noise-like signals. For spectum analyzers that do

not offer this mode, it is important to set the video
bandwidth to at least 3 times the resolution bandwidth. In
this case, use a video bandwidth of 100kHz. If the ratio of
video to resolution bandwidth is too low, the power
measurement will be inaccurate. For example, if the ratio
is 1:1, the measured power may be 0.35dB lower than the
true power. Note that the sweep time must be increased by
a factor of 10 or so in order to take advantage of the higher
number of samples in this mode. Video averaging helps
smooth the result; 100 averages gives good results. The
channel power utility of the analyzer is used to find the
total power within a 3.84MHz bandwidth.
Next measure the output spectrum of the DUT using the
same settings. For ACPR, center the measurement band
5MHz above the center of the highest carrier. To find the
true spectral regrowth power, convert the measured spectral power levels to mW and subtract the spectrum analyzer noise floor from the measured DUT power. Reconvert
to dBm to get the true spectral regrowth. Do the same for
the band 5MHz below the center of the lowest carrier. Take
the average of the two dBc figures to arrive at the average
spectral regrowth.
The ACPR/AltCPR is equal to the difference in dBc between the signal power and the spectral regrowth.
The spectrum analyzer shown also offers an ACPR measurement utility. This utility will not, however, give accurate results for highly linear devices, as it does not
compensate for the measurement system noise floor.

To measure noise at 30MHz offset, the test setup is
modified as shown in Figure 4. A tunable bandpass filter is
added to the DUT output. This filter should be set to reject
the main signal, but not attenuate the noise 30MHz above
the outermost channel. Noise measurements that do not
use this filter technique will produce degraded results for
strong RF signal levels. The reason is that the dynamic
range of the spectrum analyzer is not sufficient to accept
the power of the main signal while accurately measuring
the noise. So the main signal will tend to overload the RF
front end of the analyzer. The bandpass filter helps to
reduce the amount of signal reaching the analyzer by
approximately 20dB, while adding only about 1dB of noise
figure to the measurement frequency.
Proceeding with this measurement, measure the noise
floor of the spectrum analyzer with a 50 input termination. A narrow span of 100kHz can be used, while setting
the resolution and video bandwidths to 30kHz. The input
attenuation of the analyzer should be set to 0dB. Here
again, set the detector mode for RMS. Use the marker
noise function and video averaging to obtain a result in
dBm/Hz.
Next, connect the test setup as shown in Figure 4. Measure
the noise at 30MHz from the center of the signal frequency,
using the same settings as above. To find the true noise
level, convert the noise power and the noise floor to mW
and subtract the spectrum analizer noise floor from the
AN99-3
Application Note 99
DUT output noise power. Reconvert to dBm/Hz to get the
true noise level.
ACPR/AltCPR are observed between these extremes, where

the spectral regrowth is equal to the noise floor of the DUT.
ACPR and AltCPR vary with output signal level. For low RF
output power levels, these are limited by the output noise
floor of the DUT. At high RF output power levels, they are
determined by the linearity of the DUT. The maximum
Some sample results of these measurements are shown to

illustrate this in Figures 5 and 6. Figure 5 is a plot of ACPR,
AltCPR, and noise versus baseband drive level for a
4- carrier WCDMA signal. Figure 6 shows the same results
for a single WCDMA carrier.
145
65
56
147
67
58
149
69
151
71
ADJACENT
CHANNEL
64
ALTERNATE
CHANNEL
155
157
66
68
70
NOISE AT
30MHz OFFSET
72
81
163
83
Linear Technology Corporation
149
151
153
ALTERNATE
CHANNEL
155
157
77
161
Figure 5. LT5528 4-channel WCDMA Adjacent and Alternate

CPR and 30MHz Noise Floor Measurement vs Channel Power
AN99-4
75
159
AN99 F05
147
ADJACENT
CHANNEL
73
79
165
74
40 38 36 34 32 30 28 26 24
RF OUTPUT PER CHANNEL (dBm)
145
dBm/Hz
153
62
dBm/Hz
ACPR (dBc)
60
ACPR (dBc)
54
159
NOISE AT
30MHz OFFSET
161
163
165
85
30 28 26 24 22 20 18 16 14
RF OUTPUT PER CHANNEL (dBm)
AN99 F06
Figure 6. LT5528 1-channel WCDMA Adjacent and Alternate

CPR and 30MHz Noise Floor Measurement vs Channel Power
an99f LT/TP 0505 500 PRINTED IN USA
1630 McCarthy Blvd., Milpitas, CA 95035-7417

(408) 432-1900
FAX: (408) 434-0507 www.linear.com
LINEAR TECHNOLOGY CORPORATION 2005

Application Note 99 May 2005: Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1

Uploaded by

Copyright:

Available Formats

Application Note 99 May 2005: Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Application Note 99 May 2005: Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1

Uploaded by

Copyright:

Available Formats

Application Note 99

To measure ACPR and AltCPR, refer to the test setup

Figure 1. WCDMA ACPR Limits, Per 3GPP TS 25.104, Section 6.6.2.2.1

Figure 2. ACPR Measurement Setup

In general, the baseband source will not be ideal. It

1-channel ACPR measurement

Rejection at 3.08MHz >80dB

Rejection at 5.2MHz >20dB

2-channel ACPR measurement

Rejection at 2.6MHz >20dB

Filter part number LE1182 5M-50-720B, from TTE

An accurate measurement of the spectral regrowth of a

POWER IN 30kHz BW (dBm)

The spectrum analyzer must have a wide dynamic range.

Figure 3. ACPR Spectrum for a Single Carrier WCDMA Signal

To do this, first measure the noise floor of the spectrum

K&L 5BT SERIES

Figure 4. 30MHz Offset Noise Measurement Setup

measure noise-like signals. For spectum analyzers that do

compensate for the measurement system noise floor.

ACPR/AltCPR are observed between these extremes, where

Some sample results of these measurements are shown to

Linear Technology Corporation

Figure 5. LT5528 4-channel WCDMA Adjacent and Alternate

Figure 6. LT5528 1-channel WCDMA Adjacent and Alternate

an99f LT/TP 0505 500 PRINTED IN USA

1630 McCarthy Blvd., Milpitas, CA 95035-7417

FAX: (408) 434-0507 www.linear.com

LINEAR TECHNOLOGY CORPORATION 2005

You might also like