TDSCDMA_DnLnk_MultiCarrier_TX
This section provides parameter information for Required Parameters, Basic Parameters, Signal Parameters, and parameters for the various measurements.
Symbol

Description TD-SCDMA downlink multi-carrier TX test
Library WTB
Class TSDFTDSCDMA_DnLnk_MultiCarrier_TX
Derived From baseWTB_TX
Parameters
| Name | Description | Default | Sym | Unit | Type | Range |
|---|---|---|---|---|---|---|
| RequiredParameters | ||||||
| CE_TimeStep | Circuit envelope simulation time step | 1/1.28 MHz/16 | sec | real | (0, ∞) | |
| WTB_TimeStep | Set CE_TimeStep < = 1/1.28e6/SamplesPerChip. | |||||
| FSource | Source carrier frequency | 1900 MHz | Hz | real | (0, ∞) | |
| SourcePower | Source power | dbmtow(-20.0) | W | real | [0, ∞) | |
| FMeasurement | Measurement carrier frequency | 1900 MHz | Hz | real | (0, ∞) | |
| MeasurementInfo | Available Measurements | |||||
| PowerMeasurement | Enable power measurements? NO, YES | YES | enum | |||
| SpectrumMeasurement | Enable spectrum measurement? NO, YES | NO | enum | |||
| BasicParameters | ||||||
| SourceR | Source resistance | 50 Ohm | Ohm | real | (0, ∞) | |
| SourceTemp | Source resistor temperature | -273.15 | Celsius | real | [-273.15, ∞) | |
| EnableSourceNoise | Enable source thermal noise? NO, YES | NO | enum | |||
| MeasR | Measurement resistance | 50 Ohm | Ohm | real | [10, 1.0e6] | |
| MirrorSourceSpectrum | Mirror source spectrum about carrier? NO, YES | NO | enum | |||
| MirrorMeasSpectrum | Mirror meas spectrum about carrier? NO, YES | NO | enum | |||
| RF_MirrorFreq | Mirror source frequency for spectrum/envelope measurement? NO, YES | NO | enum | |||
| MeasMirrorFreq | Mirror meas frequency for spectrum/envelope measurement? NO, YES | NO | enum | |||
| TestBenchSeed | Random number generator seed | 1234567 | int | [0, ∞) | ||
| SignalParameters | ||||||
| GainImbalance | Gain imbalance, Q vs I | 0.0 | dB | real | (-∞, ∞) | |
| PhaseImbalance | Phase imbalance, Q vs I | 0.0 | deg | real | (-∞, ∞) | |
| I_OriginOffset | I origin offset (percent) | 0.0 | real | (-∞, ∞) | ||
| Q_OriginOffset | Q origin offset (percent) | 0.0 | real | (-∞, ∞) | ||
| IQ_Rotation | IQ rotation | 0.0 | deg | real | (-∞, ∞) | |
| SamplesPerChip | Samples per chip | 16 | S | int | [8, 32] | |
| RRC_FilterLength | RRC filter length (chips) | 16 | int | [2, 128] | ||
| ActiveTimeslot | Active Timeslot: TS0, TS2, TS3, TS4, TS5, TS6 | TS6 | enum | |||
| PowerMeasurementParameters | ||||||
| PowerDisplayPages | Power measurement display pages: | |||||
| PowerSubframes | Number of subframes averaged | 1 | int | [0, ∞) | ||
| SpectrumMeasurementParameters | ||||||
| SpecMeasDisplayPages | Spectrum measurement display pages: | |||||
| SpecMeasStart | Spectrum measurement start | 0.0 | sec | real | [0, ∞) | |
| SpecMeasStop | Spectrum measurement stop | 5.0 msec | sec | real | [0, ∞) | |
| SpecMeasSubframes | Spectrum measurement subframes | 3 | int | [0, 100] | ||
| SpecMeasResBW | Spectrum resolution bandwidth | 30 kHz | Hz | real | [0, ∞) | |
| SpecMeasWindow | Window type: none, Hamming 0.54, Hanning 0.50, Gaussian 0.75, Kaiser 7.865, _8510 6.0, Blackman, Blackman-Harris | none | enum |
Pin Inputs
| Pin | Name | Description | Signal Type |
|---|---|---|---|
| 2 | Meas_In | Test bench measurement RF input from RF circuit | timed |
Pin Outputs
| Pin | Name | Description | Signal Type |
|---|---|---|---|
| 1 | RF_Out | Test bench RF output to RF circuit | timed |
Setting Parameters
More control of the test bench can be achieved by setting parameters in the Basic Parameters, Signal Parameters, and measurement categories for the activated measurements.
| Note For required parameter information, see Set the Required Parameters. |
Basic Parameters
- SourceR is the RF output source resistance.
- SourceTemp is the RF output source resistance temperature (oC) and sets noise density in the RF output signal to (k(SourceTemp+273.15)) Watts/Hz, where k is Boltzmann's constant.
- EnableSourceNoise, when set to NO disables the SourceTemp and effectively sets it to -273.15oC (0 Kelvin). When set to YES, the noise density due to SourceTemp is enabled.
- MeasR defines the load resistance for the RF DUT output Meas signal into the test bench. This resistance loads the RF DUT output; it is also the reference resistance for Meas signal power measurements.
- MirrorSourceSpectrum is used to invert the polarity of the Q envelope of the generated RF signal before it is sent to the RF DUT. Depending on the configuration and number of mixers in an RF transmitter, the signal at the input of the DUT may need to be mirrored. If such an RF signal is desired, set this parameter to YES.
- MirrorMeasSpectrum is used to invert the polarity of the Q envelope in the Meas_in RF signal input to the test bench (and output from the RF DUT). Depending on the configuration and number of mixers in the RF DUT, the signal at its output may be mirrored compared to the signal generated by the signal source (before any mirroring is done because of the MirrorSourceSpectrum setting). Proper demodulation and measurement of the RF DUT output signal requires that its RF envelope is not mirrored compared to the signal generated by the signal source (before any mirroring is done because of the MirrorSourceSpectrum setting). If the Meas_in RF signal is mirrored, set this parameter to YES. Proper setting of this parameter is required for measurements on the Meas_in signal in TX test benches (EVM, Constellation, CDP, PCDE) and results in measurement on a signal with no spectrum mirroring.
- TestBenchSeed is an integer used to seed the random number generator used with the test bench. This value is used by all test bench random number generators, except those RF DUT components that use their own specific seed parameter. TestBenchSeed initializes the random number generation. The same seed value produces the same random results, thereby giving you predictable simulation results. To generate repeatable random output from simulation to simulation, use any positive seed value. If you want the output to be truly random, enter the seed value of 0.
- RF_MirrorFreq is used to invert the polarity of the Q envelope in the RF_out RF signal for RF envelope, ppectrum, ACLR, and occupied bandwidth measurements. RF_MirrorFreq is typically set by the user to NO when MirrorSourceSpectrum = NO; RF_MirrorFreq is typically set by the user to YES when MirrorSourceSpectrum = YES. Both settings result in viewing the RF_out signal with no spectrum mirroring. Other settings by the user result in RF_out signal for RF_Envelope and Spectrum measurements with spectrum mirroring.
- MeasMirrorFreq is used to invert the polarity of the Q envelope in the Meas_in RF signal for the RF envelope, spectrum, ACLR, and occupied bandwidth measurements.
MeasMirrorFreq is typically set to NO by the user when the combination of the MirrorSourceSpectrum value and any signal mirroring in the users RF DUT results in no spectrum mirroring in the Meas_in signal. MeasMirrorFreq is typically set to YES by the user when the combination of the MirrorSourceSpectrum and RF DUT results in spectrum mirroring in the Meas_in signal.
Other settings result in Meas_in signal for RF_Envelope and Spectrum measurements with spectrum mirroring. The MirrorMeasSpectrum parameter setting has no impact on the setting or use of the MeasMirrorFreq parameter.
Signal Parameters
- GainImbalance, PhaseImbalance, I_OriginOffset, Q_OriginOffset, and IQ_Rotation are used to add certain impairments to the ideal output RF signal. Impairments are added in the order described here.
The unimpaired RF I and Q envelope voltages have gain and phase imbalance applied. The RF is given by:

where A is a scaling factor that depends on the SourcePower and SourceR parameters specified by the user, V I ( t ) is the in-phase RF envelope, V Q ( t ) is the quadrature phase RF envelope, g is the gain imbalance

and, φ (in degrees) is the phase imbalance.
Next, the signal V RF ( t ) is rotated by IQ_Rotation degrees. The I_OriginOffset and Q_OriginOffset are then applied to the rotated signal. Note that the amounts specified are percentages with respect to the output rms voltage. The output rms voltage is given by sqrt(2 × SourceR × SourcePower). - SamplesPerChip sets the number of samples in a chip.
The default value is set to 16 to display settings according to the 3GPP NTDD. It can be set to a larger value for a simulation frequency bandwidth wider than 16 × 1.28 MHz. It can be set to a smaller value for faster simulation; however, this will result in lower signal fidelity. If SamplesPerChip = 8, the simulation RF bandwidth is larger than the signal bandwidth by a factor of 8 (e.g., simulation RF bandwidth = 8 × 1.28 MHz). - RRC_FilterLength sets root raised-cosine (RRC) filter length in chips.
The default value is set to 16 to transmit TD-SCDMA downlink signals in time and frequency domains based on the 3GPP NTDD standard. It can be set to a smaller value for faster simulation; however, this will result in lower signal fidelity. - ActiveTimeslot specifies which slot signal in the subframe will be transmitted. Referring to 12.2 kbps Downlink Channel Subframe Structure, when ActiveTimeslot=0, TS0 is used.
Power Measurement Parameters
- PowerDisplayPages provides Data Display page information for this measurement. It cannot be changed by the user.
- PowerSubframes sets the number of subframes over which data will be collected.
- The measurement start time is the time when the first subframe is detected in the measured RF signal. Automatic synchronization by the measurement model avoids any start-up transient in the Constellation plots. The measurement stop time is this start time plus PowerSubframes × SubframeTime. SubframeTime is described in Test Bench Variables for Data Displays.
Spectrum Measurement Parameters
The Spectrum measurement calculates the spectrum of the input signal. Averaging the spectrum over multiple subframes can be enabled as described in this section.
This measurement is not affected by the MirrorMeasSpectrum parameter. To apply spectrum mirroring to the measured RF_out signal, set RF_MirrorFreq = YES; to apply spectrum mirroring to the measured Meas_in signal, set MeasMirrorFreq = YES.
In the following, TimeStep denotes the simulation time step and FMeasurement denotes the measured RF signal characterization frequency.
- The measurement outputs the complex amplitude voltage values at the frequencies of the spectral tones. It does not output the power at the frequencies of the spectral tones. However, one can calculate and display the power spectrum as well as the magnitude and phase spectrum by using the dBm, mag, and phase functions of the data display window.
Note that the dBm function assumes a 50-ohm reference resistance; if a different measurement was used in the test bench, its value can be specified as a second argument to the dBm function, for example, dBm(SpecMeas, Meas_RefR) where SpecMeas is the instance name of the spectrum measurement and Meas_RefR is the resistive load used.
By default, the displayed spectrum extends from FMeasurement - 1/(2×TimeStep) Hz to FMeasurement + 1/(2×TimeStep) Hz. When FMeasurement < 1/(2×TimeStep), the default spectrum extends to negative frequencies. The spectral content at these negative frequencies is conjugated, mirrored, and added to the spectral content of the closest positive frequency. The negative frequency tones are then displayed on the positive frequency axis as would happen in an RF spectrum analyzer measurement instrument. This process may introduce an error in the displayed frequency for the mirrored tones. The absolute error introduced is less than (spectrum frequency step) / 2 (see Effect of Values for SpecMeasStart, SpecMeasStop, SpecMeasSubframes, and SpecMeasResBW for the definition of spectrum frequency step).
The basis of the algorithm used by the spectrum measurement is the chirp-Z transform. The algorithm can use multiple subframes and average the results to achieve video averaging (see note 6). - SpecMeasDisplayPages provides information regarding Data Display pages for this measurement. It cannot be changed by the user.
- SpecMeasStart sets the start time for collecting input data.
- SpecMeasStop sets the stop time for collecting input data when SpecMeasSubframes = 0 and SpecMeasResBW = 0.
- SpecMeasSubframes sets the number of segments over which data will be collected.
- SpecMeasResBW sets the resolution bandwidth of the spectrum.
Depending on the values of SpecMeasStart, SpecMeasStop, SpecMeasSubframes, and SpecMeasResBW, the stop time may be adjusted or spectrum video averaging may occur (or both). The different cases are described in Effect of Values for SpecMeasStart, SpecMeasStop, SpecMeasSubframes, and SpecMeasResBW.
Referring to Effect of Values for SpecMeasStart, SpecMeasStop, SpecMeasSubframes, and SpecMeasResBW, let
Start = TimeStep × int((SpecMeasStart/TimeStep) + 0.5)
Stop = TimeStep × int((SpecMeasStop/TimeStep) + 0.5)
(This means Start and Stop are forced to be an integer number of time step intervals.)
X = normalized equivalent noise bandwidth of the window
Start and Stop times are used for RF_out and Meas_in signal spectrum analyses. The Meas_in signal is delayed in time from the RF_out signal by the value of the RF DUT time delay. Therefore, for RF DUT time delay >0, the RF_out and Meas_in signals are inherently different and spectrum display differences can be expected.
TimeStep and SubframeTime are defined in the Test Bench Variables for Data Displays section.
Equivalent noise bandwidth (ENBW) compares the window to an ideal, rectangular filter. It is the equivalent width of a rectangular filter that passes the same amount of white noise as the window. Normalized ENBW (NENBW) is ENBW multiplied by the duration of the signal being windowed. (Refer to note 7 regarding the various window options and Window Options and Normalized Equivalent Noise Bandwidth regarding NENBW for the various windows.)Effect of Values for SpecMeasStart, SpecMeasStop, SpecMeasSubframes, and SpecMeasResBW
Case 1 SpecMeasSubframes = 0
SpecMeasResBW = 0
Resultant stop time = Stop
Resultant resolution BW = X/(Stop - Start)
Resultant spectrum frequency step = 1/(Stop-Start)
Video averaging status = NoneCase 2 SpecMeasSubframes > 0
SpecMeasResBW = 0
Resultant stop time = Start + SpecMeasSubframes x SubframeTime
For SpecMeasSubframes > 0 and SpecMeasResBW = 0
Video averaging occurs over all segment time intervals
Resultant resolution BW = X /SubframeTime
Resultant spectrum frequency step = 1/SubframeTime
Video averaging status = Yes, when SpecMeasSubframes > 1Case 3 SpecMeasSubframes = 0
SpecMeasResBW > 0
Resultant stop time = Start + N x SubframeTimeInterval
where
N = int((Stop -Start)/SubframeTimeInterval) + 1
For SpecMeasSubframes = 0 and SpecMeasResBW > 0
Define SubframeTimeInterval = TimeStep x int((X/SpecMeasResBW/TimeStep) + 0.5)
This means SubframeTimeInterval is forced to a value that is an integer number of time step intervals.
(Stop-Start) time is forced to be an integer number (N) of SubframeTimeIntervals
N has a minimum value of 1
Video averaging occurs over all SubframeTimeIntervals
Resolution bandwidth achieved is ResBW = X / SubframeTimeInterval, which is very close to SpecMeasResBW but may not be exactly the same if X/SpecMeasResBW/TimeStep is not an exact integer.
Resultant resolution BW = ResBW
Resultant spectrum frequency step = ResBW
Video averaging status = Yes when N > 1Case 4 SpecMeasSubframes > 0
SpecMeasResBW > 0
Resultant stop time = Start + M x SubframeTimeInterval
where
M = int((SpecMeasSubframes x SubframeTime)/SubframeTimeInterval) + 1
For SpecMeasSubframes > 0 and SpecMeasResBW > 0
Define SubframeTimeInterval = TimeStep x int(( X /SpecMeasResBW/TimeStep) + 0.5)
This means SubframeTimeInterval is forced to a value that is an integer number of time step intervals.
(Stop-Start) time is forced to be an integer number (M) of the SubframeTimeIntervals
M has a minimum value of 1
Video averaging occurs over all SubframeTimeIntervals
Resolution bandwidth achieved is ResBW = X / SubframeTimeInterval, which is very close to SpecMeasResBW but may not be exactly the same if X/SpecMeasResBW/TimeStep is not an exact integer.
Resultant resolution BW = ResBW
Resultant spectrum frequency step = ResBW
Video averaging status = Yes, when M > 1 - SpecMeasWindow specifies the window that will be applied to each segment before its spectrum is calculated. Different windows have different properties, affect the resolution bandwidth achieved, and affect the spectral shape. Windowing is often necessary in transform-based (chirp-Z, FFT) spectrum estimation in order to reduce spectral distortion due to discontinuous or non-harmonic signal over the measurement time interval. Without windowing, the estimated spectrum may suffer from spectral leakage that can cause misleading measurements or masking of weak signal spectral detail by spurious artifacts.
The windowing of a signal in time has the effect of changing its power. The spectrum measurement compensates for this and the spectrum is normalized so that the power contained in it is the same as the power of the input signal.
Window Type Definitions:- none

where N is the window size - Hamming 0.54

where N is the window size - Hanning 0.50

where N is the window size - Gaussian 0.75

where N is the window size - Kaiser 7.865

where N is the window size, α = N / 2, and I 0 (.) is the 0th order modified Bessel function of the first kind - 8510 6.0 (Kaiser 6.0)

where N is the window size, α = N / 2, and I 0 (.) is the 0th order modified Bessel function of the first kind - Blackman

where N is the window size - Blackman-Harris

where N is the window size.Window Options and Normalized Equivalent Noise Bandwidth
Window and Default Constant NENBW none 1 Hamming 0.54 1.363 Hanning 0.50 1.5 Gaussian 0.75 1.883 Kaiser 7.865 1.653 8510 6.0 1.467 Blackman 1.727 Blackman-Harris 2.021
- none
Simulation Measurement Displays
After running the simulation, results are displayed in the Data Display pages for each measurement activated.
| Note Measurement results from a wireless test bench have associated names that can be used in Data Display Expressions. For more information, refer to Measurement Results for Expressions for TD-SCDMA Wireless Test Benches. |
Power Measurement
The power measurement shows CCDF curves (not defined in 3GPP standards) for single carrier and multicarrier before and after DUT signals. Mean and peak power, and peak-to-average power ratios before and after the DUT are given as shown in Power Measurement Results.

Power Measurement Results
Spectrum Measurement
The spectrum measurement (not defined in 3GPP standards) shows spectrums for single- and multi-carrier signals before and after the DUT. The spectrum analyzer output contains complex amplitude voltage values and the dBm(<meas_name>, <ref_r>) expressions can be used to convert to dBm values. Spectrum for the RF and Meas signals are shown in Spectrum Measurement Results.

Spectrum Measurement Results
Test Bench Variables for Data Displays
Reference variables used to set up this test bench are listed in Test Bench Equations Derived from Test Bench Parameters and Exported to Data Display.
Test Bench Equations Derived from Test Bench Parameters and Exported to Data Display
| Data Display Parameter | Equation with Test Bench Parameters |
|---|---|
| RF_FSource | FSource |
| RF_Power_dBm | 10*log10(SourcePower)+30 |
| RF_R | SourceR |
| TimeStep | 1/(ChipRate*SamplesPerChip) |
| SubframeTime | 5 msec |
| ActiveSlot | ActiveSlotIndex |
| NumCarriers | 3 |
| NumChannelsPerCarrier | 16 |
| Meas_FMeasurement | FMeasurement |
| Meas_R | MeasR |
Baseline Performance
- Test Computer Configuration
- Pentium IV 2.4 GHz, 512 MB RAM, Red Hat Linux 7.3
- Conditions
- Measurements made with default test bench settings.
- RF DUT is an RF system behavior component.
- The number of time points in one TD-SCDMA downlink subframe is a function of SamplesPerChip and ChipRate.
- SamplesPerChip = 16
- ChipRate = 1.28 Mb/s
- Resultant WTB_TimeStep = 48.828125 nsec; SubframeTime = 5 msec; time points per subframe = 102400
TDSCDMA_DnLnk_MultiCarrier_TX Measurement Bursts Measured Simulation Time (sec) ADS Processes (MB) Power 1 144 139 Spectrum 1 301 272
Expected ADS Performance
Expected ADS performance is the combined performance of the baseline test bench and the RF DUT Circuit Envelope simulation with the same signal and number of time points. For example, if the RF DUT performance with Circuit Envelope simulation alone takes 2 hours and consumes 200 MB of memory (excluding the memory consumed by the core ADS product), then add these numbers to the Baseline Performance numbers to determine the expected ADS performance. This is valid only if the full memory consumed is from RAM. If RAM is less, larger simulation times may result due to increased disk access time for swap memory usage.
References
- 3GPP TS 25.221, "3rd Generation Partnership Project; Technical Specification Group Radio Access Network; Physical channels and mapping of transport channels onto physical channels (TDD) (Release 4)," version 4.5.0, June, 2002.
http://www.3gpp.org/ftp/Specs/2002-06/Rel-4/25_series/25221-450.zip] - 3GPP TS 25.223, "3rd Generation Partnership Project; Technical Specification Group Radio Access Network; Spreading and modulation (TDD) (Release 4)," version 4.4.0, March, 2002.
http://www.3gpp.org/ftp/Specs/2002-06/Rel-4/25_series/25223-440.zip]
Setting up a Wireless Test Bench Analysis in the Wireless Test Bench Simulation documentation explains how to use test bench windows and dialogs to perform analysis tasks.
Setting Circuit Envelope Analysis Parameters in the Wireless Test Bench Simulation documentation explains how to set up circuit envelope analysis parameters such as convergence criteria, solver selection, and initial guess.
Setting Automatic Verification Modeling Parameters in the Wireless Test Bench Simulation documentation explains how to improve simulation speed.
Privacy
Statement
|
Terms of Use
|
Legal |
Contact Us
|
© Agilent 2000-2008 ![]()