B.1 General

37.5443GPPConformance testingRelease 16TSUniversal Terrestrial Radio Access (UTRA) and Evolved UTRA (E-UTRA)User Equipment (UE) Over The Air (OTA) performance

  1. The relative power values of the measurement points will be transformed to absolute radiated power values (in dBm) by performing a calibration measurement. The calibration measurement is done by using a reference antenna with known efficiency or gain values. In the calibration measurement the reference antenna is measured in the same place as the DUT, and the attenuation of the complete transmission path () from the DUT to the measurement receiver/NB/BS simulator is calibrated out.
  2. The gain and/or radiation efficiency of the reference antenna shall be known at the frequency bands in which the calibrations are performed. Recommended calibration antennas are monopole antennas or sleeve dipoles tuned for the each frequency band of interest. Alternatively, other methods may be used if they ensure an equal or greater level of accuracy. A network analyzer is recommended to be used to perform the calibration measurement. Also other devices can be used to measure attenuation.
  3. The principle is based on the use of calibration/substitution antennas presenting an efficiency known with a sufficient accuracy in the measurement bandwidths. Such a calibration antenna is placed on the DUT positioner at the exact UE location used for TRP and TRS measurement. It is possible to use a mechanical piece to place the calibration antenna on the positioner. This mechanical piece should not present any electromagnetic properties, which could influence the frequency response and the radiation properties of the calibration antenna.

B.2 Calibration procedure

B.2.1 Calibration procedure for Anechoic chamber method

  1. For the anechoic chamber method, the calibration is performed individually for the both orthogonal polarizations, all the transmission paths and all frequencies used in the testing. An illustration of the substitution configuration is shown in Figure B.2.1-1.
  2. is the attenuation between P and B, see Figure B.2.1-1.
  3. Where is cable loss from A to C. The cable AC connecting the substitution antenna should be such that its influence upon radiation pattern measurements is minimal. is the attenuation between points A and B. In TRP and TRS measurements point B is connected to the calibrated input/output port of measurement receiver.
  4. is the efficiency or gain of the calibration antenna at the frequency of interest.
  6. Figure B.2.1-1: Calibration/substitution procedures using a vector network analyzer
  7. If the calibration is based on known efficiency of the calibration antenna, a full spherical scanning is performed to determine. Unless the otherwise specified in the calibration antenna documentation, TRP sampling grid in clause 4.4 and equation for TRP in clause 6 should be used for head and hand phantom and TRP sampling grid in clause 4.4 and equation for TRP in clause 6 should be used for laptop ground plane phantom and embedded devices.
  8. This procedure has to be done at each frequency of interest.
  9. To achieve measurements with an uncertainty as low as possible, it is absolutely necessary to exactly keep the same P to B configuration (cables, dual-polarized antenna.

B.2.2 Calibration procedure for Reverberation chamber method

For the reverberation chamber method, all polarizations and transmission paths are included in one calibration measurement. The calibration antenna can be place in an arbitrarily position, as long as it is placed 0.5 wavelengths from other metallic objects and 0.7 wavelengths from absorbing objects. An illustration of the measurement setup for this procedure is shown in Figure B.2.2-1.

  1. The purpose of the calibration measurement is to determine the average power transfer function in the chamber, mismatch of fixed measurement antennas and path losses in cables connecting the power sampling instrument and the fixed measurement antennas. Preferably a network analyzer is used for these measurements. Recommended calibration antennas are dipoles tuned to the frequency band of interest.
  2. In general, the calibration of a reverberation chamber is performed in three steps:
    1. 1. Measurement of S-parameters through the reverberation chamber for a complete stirring sequence
  3. 2. Calculation of the chamber reference transfer function
  4. 3. Measurement of connecting cable insertion loss
  5. If several setups are used (e.g. empty chamber, chamber with head or hand phantom, etc.), steps 1 and 2 must be repeated for each configuration. The calibration measurement setup is shown in Figure B.2.2-1.
  7. Figure B.2.2-1: Calibration measurement setup in the reverberation chamber, using a vector network analyzer

B.2.2.1 Measurement of S-parameters through the chamber for a complete stirring sequence

  1. This step will measure S-parameters through the reverberation chamber through a complete stirring sequence. This information is required to determine the chamber’s reference transfer function. The procedure must be performed separately for each measurement setup of which the loading of the chamber has been changed. The calibration procedure must be repeated for each frequency as defined above. Therefore, it is advantageous if the network analyzer can be set to a frequency sweep covering the defined frequencies, so that all frequencies of interest can be measured with a minimal number of measurement runs.
  2. i. Place all objects into the RC which will be used during TRP or TRS measurements, including a head phantom, hand phantom and fixture for the EUT. This ensures that the loss in the chamber, which determines the average power transfer level, is the same during both calibration and test measurements. Also, if the EUT is large or contains many antennas, it may represent a noticeable loading of the chamber. It should then be present in the chamber and turned on during the calibration.
  3. ii. Place the calibration antenna inside the chamber. The calibration antenna is preferably mounted on a low-loss dielectric fixture, to avoid effects from the fixture itself which may affect the EUT’s radiation efficiency and mismatch factor. The calibration antenna must be placed in the chamber in such a way that it is far enough from any walls, mode-stirrers, head phantom, hand phantom, or other object, such that the environment for the calibration antenna (taken over the complete stirring sequence) resembles a free space environment. “Far enough away” depends on the type of calibration antenna used. For low gain nearly omni-directional antennas like dipoles, it is normally sufficient to ensure that this spacing is larger than 0.5 wavelengths. More directive calibration antennas should be situated towards the centre of the chamber. The calibration antenna should be present in the chamber during the TRP/TRS measurements.
  4. iii. Calibrate the network analyzer with a full 2-port calibration in such a way that the vector S-parameters between the ports of the fixed measurement antenna and the calibration antenna can be accurately measured. Preferably, the network analyzer is set to perform a frequency sweep at each stirrer position. This will enable calibration of several frequency points during the same stirring sequence, thereby reducing calibration time. This will also enable frequency stirring, i.e., averaging the measured power transfer function over a small frequency bandwidth around each measured frequency point (moving frequency window). This will increase accuracy at the expense of frequency resolution.
  5. iv. Connect the antennas and measure the S-parameters for each stirrer position and each fixed measurement antenna.
  6. The number of stirrer positions in the chosen stirring sequence, i.e. the number of S-parameter samples at each frequency point, should be chosen in such a way that it is large enough to yield an acceptable statistical contribution to the total measurement uncertainty. As a guideline it should be larger than 100, preferable 200 or 400 to ensure that the number of independent samples is not severely limited by the total number of samples measured. The number of independent samples, which is a subset of all samples, determines the statistical contribution to the expanded accuracy (which is two times the standard deviation). This should be not less than 100 to ensure an expanded accuracy better than 0.5 dB. The number of independent samples depends on the operating frequency, volume of the chamber, efficacy of the chamber’s stirrers, the level of loading by absorbing objects, and whether or not frequency stirring is used.
  7. The sequence of moving the stirrers to different positions may be either step-wise (stopping stirrer for each sample) or continuous (sampling on-the-fly). With continuous stirring it may not be possible to characterize the chamber over a wide frequency band at the same time.

B.2.2.2 Calculation of the chamber reference transfer function

From the S-parameters obtained in the calibration measurement, the chamber reference transfer function for fixed antenna n can be calculated. The reflection coefficient for fixed antenna n can be calculated by the following equation,

Thus, the chamber reference transfer function can be calculated by the following equation,

where M is the total number of samples of the transfer function measured for each fixed measurement antenna and is sample number m of the transfer function for measurement antenna n. Moreover, is the complex average of the calibration antenna reflection coefficient. Finally, is the radiation efficiency of the calibration antenna.

NOTE: The radiation efficiency of the fixed antenna is not corrected for because it will be the same both during calibration and measurements. Therefore, the fixed antenna’s radiation efficiency will not affect the final results. The same can be said about the mismatch factor of the fixed measurement antennas, but it is still advantageous to correct for this factor if frequency stirring is applied to improve accuracy.

B.2.2.3 Cable calibration

This measurement step will calibrate the power loss of the cable needed to connect the instrument used to measure the received power at the fixed measurement antenna during TRP measurements, and to generate the power radiated by the fixed antenna during TRS measurements.

i. Disconnect the cables between the VNA and the chamber.

ii. Connect the cables one-by-one between the two ports of the network analyzer. The VNA must be calibrated at its own two ports.

iii. Measure the frequency response of the transmission S-parameter ( or ) of the cable.

iv. Save the power transfer values () of the frequency response curve for the test frequencies and cables positions, etc).

Calibration shall be performed yearly or if any equipment in the measurement system is changed.

B.2.3 Calibration procedure for Anechoic chamber method with Multiprobe configuration

The system needs to be calibrated in two steps in order to ensure that the absolute power is correct. The first calibration steps ensures the accurate generation of the channel model in the centre of the chamber as required by Annex L.4.
The second step validates the total power as would be seen by the DUT and allows for that power to be scaled up or down if necessary.

Considering the complexity of the system various way to calibrate are possible. The end goals are however the same no matter the exact procedure. The two steps must achieve the following:

Step 1: This step is used to equalize the power in the centre coming from the different probes. This being a relative measurement is very robust and with minimal uncertainty. It is sufficient to use instruments calibrated according to the manufacturer’s specifications and the measurements require no additional calibration. This step is done for both vertical and horizontal polarizations. The relative differences between probe path losses are recorded and used (typically in the fading emulator) to adjust the generated fading signals for each probe. Example measurement set-up is shown in Figure B.2.3.1-1.

NOTE 1: If Step 1 is performed as an absolute measurement accounting for the cable and reference antenna gains Step 2 can be omitted.

Step 2: This step is used to measure the total absolute power of at least one polarization in the centre of the ring. Then assuming that validation of the channel models has been done, the total power available to the DUT in the centre of the chamber can be computed. If necessary the power can be scaled up or down to achieve the desired power level. Since this is an absolute power measurement, the measurement cable and reference antenna gains have to be accounted for.

NOTE 2: To minimize measurement uncertainty the passive and active components of the system may be calibrated independently as well as at different intervals.

NOTE 3: Step 2 of the calibration should be performed with the channel model loaded and LTE signalling active. Sufficient amount of time averaging is required because of the fading nature of the models used.

NOTE 4: Various ways of performing the two steps may exist depending on the equipment used.
The lab is responsible for providing a comprehensive calibration procedure.

NOTE 5: Steps one and two may be combined with the channel verification procedure.

NOTE 6: The calibration must be performed for all frequencies of interest.

B.2.3.1 Example Calibration Procedure

The calibration procedure outlined below is only one possibility based on a concrete measurement set-up. Improvements can be made to minimize measurement uncertainty.

Step 1 (see Figure B.2.3.1-1)

1. Place a vertical reference dipole in the centre of the chamber, connected to a VNA port, with the other VNA port connected to the input of the channel emulator unit.

2. Configure the channel emulator for bypass mode (NOTE this might not be available in all instruments)

3. Measure the response of each path from each vertical polarization probe to the reference antenna in the centre.

4. Adjust the power on all vertical polarization branches of the channel emulator so that the powers received at the centre are equal.

5. Repeat the steps 1 to 4 with the magnetic loop and horizontally polarized probes instead, and adjust the horizontal polarization branches of the channel emulator.

NOTE: At this stage all vertical polarization paths have equalized gains, and so do all horizontal polarization paths. The two polarizations however do not necessarily produce the same signal strength in the centre of the chamber – this most commonly happens if two physically different channel emulators are used for the two polarizations. The resulting power imbalance can be accounted for either at this step or adjusted at point 7 of step 2.

Figure B.2.3.1-1: Setup for VNA measurements

Step 2 (see Figure B.2.3.1-2)

1. Place a vertical reference dipole in the centre of the chamber connected to a spectrum analyzer via an RF cable. NOTE: A power meter can also be used.

2. Record the cable and reference dipole gains.

3. Load the target channel model

4. Start the LTE signalling in the base station emulator with the required parameter identical to the measurements conditions (some special instrument options might be necessary).

5. Average the power received by the spectrum analyzer for a sufficient amount of time to account for the fading channel – one full channel simulation might be unnecessary.

6. Repeat steps 1 to 4 with a magnetic loop for the horizontal polarization (NOTE: this way no prior validation of the channel model is required)

7. Calculate the total power received at the test area as the sum of the power in the two polarizations.

8. Adjust the power in the two polarizations if necessary. The power adjustment can be a simple scaling of the power up or down or adjustment of the XPR due to slight differences in the fading unit’s branches. Depending on the adjustment needed, it can be done at the base station emulator or the channel emulator or both.

Figure B.2.3.1-2: Example setup for step 2 of the calibration

B.2.4 Calibration procedure for RTS method

For the RTS method, only the DUT reporting RSAP calibration is to perform. The RTS method depends on reported RSAP to do the antenna pattern measurement, and the power calibration for radiated second-stage throughput test. Since these reported readings do not come from a calibrated measurement instrument, their accuracies are often subject to questioning. Below procedure provides method for proper calibration to validate the reported RSAP accuracy.

– DUT reporting RSAP calibration:

With proper spherical coordinate definition of mobile terminal setup, the receive antenna pattern at any coordinate (θi, ϕi) can be expressed as:

P(θi, ϕi)= RSi (1)

The test point (θi, ϕi) can be at any point on the 4π solid angle of the coordinate. More precisely, the reported RS of the UE can be expressed as:

RSi (x)= m(x)*x+c (2)

where m(xi) and c is a function of signal strength independent of testing point angular coordinates. x is the actual incident field power density that can be derived from the signal power and test range loss. The above equation (2) assumes that the signal variables in both sides of the equation are expressed in decibels or dB. When m(x)=1 and c=g0, where g0 is the received antenna gain at the test point, we would have declared that the RS report is a true reading of the signal strength. But in reality, m(x) can be biased by either the signal level relative to the receiver’s detector operating condition, or by application software programming errors. Meanwhile, the offset constant c can also be biased by either the noise floor of the receiver and/or other artificial factors in the UE RS reporting. Therefore, a Taylor’s series can be introduced to have a better representation of the RS report value:

RSi (x)= c +ax+ bx2 +dx3 + ex4 + …. (3)

In theory, equation (3) may require many terms to represent the RS report accurately. However, since the reported RS reading in the receive antenna pattern in (1) has a limited signal dynamic range, the following three term expression is enough to correct for the reporting errors:

RSi (x)= c +ax+ bx2 +o(x3) (4)

The third order term o(x3) is ignored since it is not significant for limited signal dynamic range.

The RS pattern in (1) will need to be calibrated for its possible reporting errors. One practical method is to introduce signal correction terms in equation (4) to enhance the accuracy. This process is also referred to as the linearization. The following steps describe what is needed to linearize the RS report and the RS antenna pattern in (1):

1. Find the maximum RS reading in the pattern obtained in (1) from all angular test points, and at all polarizations. Set this point as the reference point and record signal generator power P0 and let x0 =P0-Test Range Path Loss (in dB), and note the maximum RS reading as r0.

2. At the reference test point, decrease the signal generator’s output power Pi, and let xi = Pi-Test Range Path Loss (in dB), starting from P0 with a step size (1.0 dB by default) to a power range so as to obtain the full RS reading range while searching from the reference point r0 when the test system dynamic range allows, or to 20 dB by default when the system dynamic range is known to be limited, whichever is lower in range. Record the corresponding RS reading of the UE as ri. Repeat this process until all angular test points are completed as required.

3. Use those pairs of data obtained in Step 2) (ri, xi) as the input of the quadratic fitting curve in (4) to formulate the Algorithm LSF (Least Square Fit) to calculate the three coefficients (a,b, and c) in:

Err(a,b,c)= Σ(c + axi + bxi – ri)2 (5)

Please note that the power range of the linearization can be limited to 20 dB as stated in Step 2) since the lower RS readings than the processed range do not contribute to the overall receive antenna pattern as much.

Once the three coefficients (a,b, and c) are obtained in LSF of (5), the use of the inverse function of Equation (4) to convert RS pattern in (1) into the normalized incident power pattern. This process completes the error correction for both constant biased and the non-linearity in the RS report from the UE within the limited signal dynamic range as tested in (1) and Step 1. Similar steps can be applied for error corrections if multiple receiver RS are involved in the testing.

Of course, the reference test point can also be further tested for the threshold of the receiver sensitivity and/or throughput knee-point for the required calibrated reference power level.

Annex C (informative): Measurement test report