6.4 Transmit signal quality
38.101-13GPPNRPart 1: Range 1 StandaloneRelease 17TSUser Equipment (UE) radio transmission and reception
6.4.1 Frequency error
The UE basic measurement interval of modulated carrier frequency is 1 UL slot. The mean value of basic measurements of UE modulated carrier frequency shall be accurate to within ± 0.1 PPM observed over a period of 1 ms of cumulated measurement intervals compared to the carrier frequency received from the NR Node B.
6.4.2 Transmit modulation quality
6.4.2.0 General
Transmit modulation quality defines the modulation quality for expected in-channel RF transmissions from the UE. The transmit modulation quality is specified in terms of:
– Error Vector Magnitude (EVM) for the allocated resource blocks (RBs)
– EVM equalizer spectrum flatness derived from the equalizer coefficients generated by the EVM measurement process
– Carrier leakage
– In-band emissions for the non-allocated RB
All the parameters defined in clause 6.4.2 are defined using the measurement methodology specified in Annex F.
In case the parameter 3300 or 3301 is reported from UE via the parameter txDirectCurrentLocation in UplinkTxDirectCurrentList IE (as defined in TS 38.331 [7]), carrier leakage measurement requirement in clause 6.4.2.2 and 6.4.2.3 shall be waived, and the RF correction with regard to the carrier leakage and IQ image shall be omitted during the calculation of transmit modulation quality.
6.4.2.1 Error Vector Magnitude
The Error Vector Magnitude is a measure of the difference between the reference waveform and the measured waveform. This difference is called the error vector. Before calculating the EVM the measured waveform is corrected by the sample timing offset and RF frequency offset. Then the carrier leakage shall be removed from the measured waveform before calculating the EVM.
The measured waveform is further equalised using the channel estimates subjected to the EVM equaliser spectrum flatness requirement specified in clause 6.4.2.4. For DFT-s-OFDM waveforms, the EVM result is defined after the front-end FFT and IDFT as the square root of the ratio of the mean error vector power to the mean reference power expressed as a %. For CP-OFDM waveforms, the EVM result is defined after the front-end FFT as the square root of the ratio of the mean error vector power to the mean reference power expressed as a %.
The basic EVM measurement interval in the time domain is one preamble sequence for the PRACH and one slotfor PUCCH and PUSCH in the time domain. The EVM measurement interval is reduced by any symbols that contains an allowable power transient in the measurement interval, as defined in clause 6.3.3.
The RMS average of the basic EVM measurements over 10 subframes for the average EVM case, and over 60 subframes for the reference signal EVM case, for the different modulation schemes shall not exceed the values specified in Table 6.4.2.1-1 for the parameters defined in Table 6.4.2.1-2. For EVM evaluation purposes, all 13 PRACH preamble formats and all 5 PUCCH formats are considered to have the same EVM requirement as QPSK modulated..
Table 6.4.2.1-1: Requirements for Error Vector Magnitude
|
Parameter |
Unit |
Average EVM Level |
|
Pi/2-BPSK |
% |
30 |
|
QPSK |
% |
17.5 |
|
16 QAM |
% |
12.5 |
|
64 QAM |
% |
8 |
|
256 QAM |
% |
3.5 |
Table 6.4.2.1-2: Parameters for Error Vector Magnitude
|
Parameter |
Unit |
Level |
|
UE Output Power |
dBm |
≥ Table 6.3.1-1 |
|
UE Output Power for 256 QAM |
dBm |
≥ Table 6.3.1-1 + 10 dB |
|
Operating conditions |
Normal conditions |
6.4.2.1a Error Vector Magnitude including symbols with transient period
In 6.4.2.1, EVM has been defined by excluding the symbols which have a transient period. In this section, measurement interval is defined for the symbols with a transient period to include these symbols in the RMS average EVM computation when the UE reports a transient period capability other than the default. Before calculating the EVM, the measured waveform is corrected for sample timing offset and RF frequency offset. Then the carrier leakage shall be removed from the measured waveform before calculating the EVM. The symbols with transient period should not be used for equalization. Only CP-OFDM waveform is used for conformance testing.
In the case of PUSCH or PUCCH transmissions when the mean power, modulation or RB allocation across slot or subslot boundaries is expected to change the EVM result over the symbols where the transient occurs is calculated according to Table 6.4.2.1a-1.
Table 6.4.2.1a-1: EVM definition for reported transient period
|
Reported transient capability (us) |
EVM definition |
tpstart (µs) |
SCS4 |
|---|---|---|---|
|
2 |
-0.5 |
15kHz or 30kHz5 |
|
|
4 |
-1 |
15kHz |
|
|
7 |
-2.7 |
15kHz |
|
|
NOTE 1: ,,and are defined in Annex F NOTE 2: is the EVM for a symbol right after a transition; is the EVM for a symbol right before a transition NOTE 3: tpstart denotes the start position of the EVM exclusion window as shown in Annex F.4 NOTE 4: SCS denotes the SCS that can be used in the conformance test NOTE 5: 30kHz shall be used in the conformance test unless the UE signals in supportedSubCarrierSpacingUL in FeatureSetPerCC that it only supports 15kHz in the corresponding band |
|||
The RMS average of the basic EVM measurements over 108 subframes for the symbols where the transient occurs for the different modulation schemes shall not exceed the values specified in Table 6.4.2.1a-2 for the parameters defined in Table 6.4.2.1a-3. This requirement can be verified with 64 QAM and 256 QAM modulation.
Table 6.4.2.1a-2: Requirements for Error Vector Magnitude
|
Parameter |
Unit |
Average EVM Level |
|
64 QAM |
% |
10 |
|
256 QAM |
% |
8 |
Table 6.4.2.1a-3: Parameters for Error Vector Magnitude
|
Parameter |
Unit |
Level |
|
UE Output Power |
dBm |
≥ Table 6.3.1-1 |
|
UE Output Power for 256 QAM |
dBm |
≥ Table 6.3.1-1 + 10 dB |
|
Operating conditions |
Normal conditions |
6.4.2.2 Carrier leakage
Carrier leakage is an additive sinusoid waveform whose frequency is the same as the modulated waveform carrier frequency. The measurement interval is one slot in the time domain.
In the case that uplink sharing, the carrier leakage may have 7.5 kHz shift with the carrier frequency.
The relative carrier leakage power is a power ratio of the additive sinusoid waveform and the modulated waveform. The relative carrier leakage power shall not exceed the values specified in Table 6.4.2.2-1.
Table 6.4.2.2-1: Requirements for Carrier Leakage
|
Parameter |
Relative Limit (dBc) |
|
Output power > 10 dBm |
-28 |
|
0 dBm ≤ Output power ≤ 10 dBm |
-25 |
|
-30 dBm ≤ Output power < 0 dBm |
-20 |
|
-40 dBm ≤ Output power < -30 dBm |
-10 |
6.4.2.3 In-band emissions
The in-band emission is defined as the average emission across 12 sub-carriers and as a function of the RB offset from the edge of the allocated UL transmission bandwidth. The in-band emission is measured as the ratio of the UE output power in a non–allocated RB to the UE output power in an allocated RB.
The basic in-band emissions measurement interval is defined over one slot in the time domain; however, the minimum requirement applies when the in-band emission measurement is averaged over 10 sub-frames. When the PUSCH or PUCCH transmission slot is shortened due to multiplexing with SRS, the in-band emissions measurement interval is reduced by one or more symbols, accordingly.
The average of the basic in-band emission measurement over 10 sub-frames shall not exceed the values specified in Table 6.4.2.3-1.
Table 6.4.2.3-1: Requirements for in-band emissions
|
Parameter description |
Unit |
Limit (NOTE 1) |
Applicable Frequencies |
|
|
General |
dB |
|
Any non-allocated (NOTE 2) |
|
|
IQ Image |
dB |
-28 |
Image frequencies when output power > 10 dBm |
Image frequencies (NOTES 2, 3) |
|
-25 |
Image frequencies when output power ≤ 10 dBm |
|||
|
Carrier leakage |
dBc |
-28 |
Output power > 10 dBm |
Carrier leakage frequency (NOTES 4, 5) |
|
-25 |
0 dBm ≤ Output power ≤ 10 dBm |
|||
|
-20 |
-30 dBm ≤ Output power < 0 dBm |
|||
|
-10 |
-40 dBm ≤ Output power < -30 dBm |
|||
|
NOTE 1: An in-band emissions combined limit is evaluated in each non-allocated RB. For each such RB, the minimum requirement is calculated as the higher of – 30 dB and the power sum of all limit values (General, IQ Image or Carrier leakage) that apply. is defined in NOTE 10. NOTE 2: The measurement bandwidth is 1 RB and the limit is expressed as a ratio of measured power in one non-allocated RB to the measured average power per allocated RB, where the averaging is done across all allocated RBs. For pi/2 BPSK with Spectrum Shaping, the limit is expressed as a ratio of measured power in one non-allocated RB to the measured power in the allocated RB with highest PSD. NOTE 3: The applicable frequencies for this limit are those that are enclosed in the reflection of the allocated bandwidth, based on symmetry with respect to the carrier leakage frequency, but excluding any allocated RBs. NOTE 4: The measurement bandwidth is 1 RB and the limit is expressed as a ratio of measured power in one non-allocated RB to the measured total power in all allocated RBs. NOTE 5: The applicable frequencies for this limit depend on the parameter txDirectCurrentLocation in UplinkTxDirectCurrent IE, and are those that are enclosed either in the RB containing the carrier leakage frequency, or in the two RBs immediately adjacent to the carrier leakage frequency but excluding any allocated RB. NOTE 6: LCRB is the Transmission Bandwidth (see clause 5.3). NOTE 7: NRB is the Transmission Bandwidth Configuration (see clause 5.3). NOTE 8: EVM is the limit specified in Table 6.4.2.1-1 for the modulation format used in the allocated RBs. NOTE 9: NOTE 10: NOTE 11: For almost contiguous allocations defined in clause 6.2.2, LCRB = NRB_alloc + NRB_gap with no in-gap emission requirement. |
||||
is the starting frequency offset between the allocated RB and the measured non-allocated RB (e.g. ∆RB = 1 or ∆RB = -1 for the first adjacent RB outside of the allocated bandwidth.
is an average of the transmitted power over 10 sub-frames normalized by the number of allocated RBs, measured in dBm. 6.4.2.4 EVM equalizer spectrum flatness
The zero-forcing equalizer correction applied in the EVM measurement process (as described in Annex F) must meet a spectral flatness requirement for the EVM measurement to be valid. The EVM equalizer spectrum flatness is defined in terms of the maximum peak-to-peak ripple of the equalizer coefficients (dB) across the allocated uplink block. The basic measurement interval is the same as for EVM.
The peak-to-peak variation of the EVM equalizer coefficients contained within the frequency range of the uplink allocation shall not exceed the maximum ripple specified in Table 6.4.2.4-1 for normal conditions. For uplink allocations contained within both Range 1 and Range 2, the coefficients evaluated within each of these frequency ranges shall meet the corresponding ripple requirement and the following additional requirement: the relative difference between the maximum coefficient in Range 1 and the minimum coefficient in Range 2 must not be larger than 5 dB, and the relative difference between the maximum coefficient in Range 2 and the minimum coefficient in Range 1 must not be larger than 7 dB (see Figure 6.4.2.4-1).
The EVM equalizer spectral flatness shall not exceed the values specified in Table 6.4.2.4-2 for extreme conditions. For uplink allocations contained within both Range 1 and Range 2, the coefficients evaluated within each of these frequency ranges shall meet the corresponding ripple requirement and the following additional requirement: the relative difference between the maximum coefficient in Range 1 and the minimum coefficient in Range 2 must not be larger than 6 dB, and the relative difference between the maximum coefficient in Range 2 and the minimum coefficient in Range 1 must not be larger than 10 dB (see Figure 6.4.2.4-1).
Table 6.4.2.4-1: Requirements for EVM equalizer spectrum flatness (normal conditions)
|
Frequency range |
Maximum ripple (dB) |
|
FUL_Meas – FUL_Low ≥ 3 MHz and FUL_High – FUL_Meas ≥ 3 MHz (Range 1) |
4 (p-p) |
|
FUL_Meas – FUL_Low < 3 MHz or FUL_High – FUL_Meas < 3 MHz (Range 2) |
8 (p-p) |
|
NOTE 1: FUL_Meas refers to the sub-carrier frequency for which the equalizer coefficient is evaluated NOTE 2: FUL_Low and FUL_High refer to each NR frequency band specified in Table 5.2-1 |
|
Table 6.4.2.4-2: Minimum requirements for EVM equalizer spectrum flatness (extreme conditions)
|
Frequency range |
Maximum Ripple (dB) |
|
FUL_Meas – FUL_Low ≥ 5 MHz and FUL_High – FUL_Meas ≥ 5 MHz (Range 1) |
4 (p-p) |
|
FUL_Meas – FUL_Low < 5 MHz or FUL_High – FUL_Meas < 5 MHz (Range 2) |
12 (p-p) |
|
NOTE 1: FUL_Meas refers to the sub-carrier frequency for which the equalizer coefficient is evaluated NOTE 2: FUL_Low and FUL_High refer to each NR frequency band specified in Table 5.2-1 |
|
f
FUL_High
FUL_High – 3(5) MHz
< 4(4) dBp-p
Range 1
Range 2
max(Range 1)-min(Range 2) < 5(6) dB
max(Range 2)-min(Range 1) < 7(10) dB
< 8(12) dBp-p
Figure 6.4.2.4-1: The limits for EVM equalizer spectral flatness with the maximum allowed variation of the coefficients indicated (the ETC minimum requirement are within brackets).
6.4.2.4.1 Requirements for Pi/2 BPSK modulation
These requirements apply if the IE powerBoostPi2BPSK is set to 1 for power class 3 capable UE operating in TDD bands n40, n41, n77, n78 and n79 with Pi/2 BPSK modulation and UE indicates support for UE capability powerBoosting-pi2BPSK and 40 % or less slots in radio frame are used for UL transmission. These requirements also apply if the IE dmrs-UplinkTransformPrecoding-r16 is configured and UE indicates support for UE capability lowPAPR-DMRS-PUSCHwithPrecoding-r16. Otherwise the requirements for EVM equalizer spectrum flatness defined in clause 6.4.2.4 apply
The EVM equalizer coefficients across the allocated uplink block shall be modified to fit inside the mask specified in Table 6.4.2.4.1-1 for normal conditions, prior to the calculation of EVM. The limiting mask shall be placed to minimize the change in equalizer coefficients in a sum of squares sense.
Table 6.4.2.4.1-1: Mask for EVM equalizer coefficients for Pi/2 BPSK, normal conditions
|
Frequency range |
Parameter |
Maximum ripple (dB) |
|
|FUL_Meas – Fcenter| ≤ X MHz (Range 1) |
X1 |
6 (p-p) |
|
|FUL_Meas – Fcenter| > X MHz (Range 2) |
X2 |
14 (p-p) |
|
NOTE 1: FUL_Meas refers to the sub-carrier frequency for which the equalizer coefficient is evaluated NOTE 2: Fcenter refers to the center frequency of an allocated block of PRBs NOTE 3: X, in MHz, is equal to 25% of the bandwidth of the PRB allocation NOTE 4: See Figure 6.4.2.4.1-1 for description of X1, X2 |
||
Figure 6.4.2.4.1-1: The limits for EVM equalizer spectral flatness with the maximum allowed variation. .
For Pi/2 BPSK modulation the UE shall be allowed to employ spectral shaping and the shaping filter shall be restricted so that the impulse response of the shaping filter itself shall meet
│ãt(t,0)│ ≥ │ãt(t, τ)│ ∀τ ≠ 0
20log10│ãt(t,τ)│< -15 dB 1< τ < M – 1,
where│ãt(t, τ)│=IDFT{│ãt(t,f)│ejφ (t,f)}, f is the frequency of the M allocated subcarriers , ã(t,f) and φ(t,f) are the amplitude and phase response.
0 dB reference is defined as 20log10│ãt(t,0)│.
6.4.2.5 Phase continuity requirements for DMRS bundling
For bands that UE indicates the support of DMRS bundling, the maximum allowable difference between the measured phase value in any slot p-1 and slot p, or slot 0 and any slot p for each antenna connector shall satisfy the requirements as listed in Table 6.4.2.5-1 for the measurement conditions defined in Table 6.4.2.5-2, within a measurement time window limited by the UE capability of maximum duration for DMRS bundling [maxDurationDMRS-Bundling-r17], and defined for each frequency band separately. The phase value for each slot is measured as shown in Annex F.9. These requirements apply to PUCCH and PUSCH transmissions with DFT-s-OFDM and CP-OFDM waveforms.
Table 6.4.2.5-1: Maximum allowable phase difference for DMRS bundling
|
UL channel |
Modulation order |
Phase difference between any slot p-1 and slot p (NOTE 2) |
Phase difference between slot 0 and any slot p (NOTE 3) |
|
PUSCH |
Pi/2 BPSK, QPSK |
[25] degrees |
[30] degrees |
|
PUCCH |
Pi/2 BPSK, BPSK, QPSK |
||
|
NOTE 1: The UE capability of the length of maximum duration refers to the maximum time duration during which UE is able to meet the phase continuity requirements, assuming no phase consistency violating events defined in TS 38.214 in between. NOTE 2: This requirement applies for FDD and TDD bands, for supported DMRS bundling configurations ≤ 8 slots. NOTE 3: This requirement applies only for FDD bands, for supported DMRS bundling configurations of 16 slots. |
|||
The above requirements are applicable when all the following conditions are met within the measurement time window:
– RB allocation in terms of length and frequency position does not change, and intra-slot and inter-slot frequency hopping is not activated.
– Modulation order does not change.
– No network commanded TA takes effect.
– The TPMI precoder does not change.
– There is no change in UE transmission power level, and no change in the level of P-MPR applied by the UE.
– UE is not scheduled with uplink transmission of other physical channel/signal in-between the PUSCH or PUCCH transmissions.
– For TDD, no downlink slot(s) or downlink symbol(s) or flexible symbol(s) with/without DL monitoring occasion configured in-between the PUSCH or PUCCH transmissions.
Table 6.4.2.5-2: Measurement conditions for the maximum allowable phase difference
|
Parameter |
Unit |
Level |
|
UE Output Power |
dBm |
PCMAX,f,c in clause 6.2.4, P-MPR = 0 |
|
UE downlink received power |
Not change |
|
|
Operating conditions |
Normal conditions |
|
|
Transmission bandwidth |
Confined within FUL_low + [4] MHz and FUL_high – [4] MHz |
|
|
DL signal frequency |
Not change before and during the measurement window |
|
|
DL signal timing |
Maintained constant before and during the measurement window |
|
|
UL slots for testing |
Tested on consecutive UL slots |
|
|
PUSCH waveform for testing |
DFT-s-OFDM |