Difference between revisions of "LTE UL and DL Power Measurements"

From YateBTS
Jump to: navigation, search
(Setting fixed UL power on PUSCH)
(updated for new revision)
Line 1: Line 1:
This page describes UL and DL power measurements that can be made with the YateENB.
+
This page describes UL and DL power and path loss measurements that can be made with the YateENB, as of r7472.
  
== UL Power ==
+
== UL power-related measurements ==
 +
The ENB can provide real-time measurements of UL transmitted power, received power, and path loss.
 +
 
 +
* Transmitted power is calculated from http://www.sharetechnote.com/html/Handbook_LTE_PHR.html reported power headroom], assuming that the UE is in the standard power class, with a maximum output of 23 dBm.
 +
* RSSI is calculated based on the known receiver gain for the radio hardware.
 +
* Path loss is calculated from transmitter power and RSSI. This is a port-to-port path loss, where all antennas and cables are included in the measured "path".
  
 
=== Setting fixed UL power on PUSCH ===
 
=== Setting fixed UL power on PUSCH ===
In normal operation, the ENB uses dynamic power control to adjust the output power of the UE so that the RSSI at the ENB receiver meets a target level.
+
The ENB uses dynamic power control to adjust the output power of the UE so that the ''per-resource-block'' RSSI at the ENB receiver meets a configured target level.
 +
Because the number of resource blocks in an UL transmission vary depending on load, this means that the total transmitted power of the UE changes always even though the power per resource block is fixed.
  
To make path loss and power measurements easier, the ENB supports fixed UL power as a special test mode.
+
To make path loss and power measurements easier from outside the ENB, the ENB supports fixed UL assignment sizes as a special test mode.
 +
This mode does not ''really'' fix the UL power to a constant level, but if the path loss is constant, UL power on PUSCH will also be constant.
  
To enable fixed UL power, set the PuschFixedPower configuration parameter to true:
+
To enable fixed UL assignment sizes, set the PuschFixedPower configuration parameter to true:
 
<pre>
 
<pre>
 
[basic]
 
[basic]
; Fixed UL power
+
; Fixed UL grant sizes for easier power measurement.
; If true, the output power is p0_NominalPUSCH
+
; BE CAREFUL in direct-wired connections. The UE might have enough power to damage the ENB receiver.
+
 
PuschFixedPower=true
 
PuschFixedPower=true
 
</pre>
 
</pre>
Line 20: Line 25:
 
(If you do not already have such a file, just create one. ''Be sure to indicate the "[basic]" section at the start of the file.'')
 
(If you do not already have such a file, just create one. ''Be sure to indicate the "[basic]" section at the start of the file.'')
  
In fixed UL power mode, the output power of the UE is controlled by p0_NominalPUSCH, which ''is'' available in the LMI, in the "Access Channels : PUSCH" page.
+
The default size for the fixed UL grant is 4 resource blocks.
So if you set PuschFixedPower, be sure to check this value also.
+
  
The absolute output power of the UE will be p0_NominalPUSCH in dBm ''per resource block''.
+
'''Note: This mode is not necessary for UL power, RSSI and power loss measurements.'''
To maintain fixed power, '''the ENB will fix all PUSCH allocations at 6 resource blocks'''.
+
This severely limits UL data throughput, but gives a consistent power level on PUSCH.
+
  
With the fixed allocation of 6 resource blocks, the PUSCH total power is
+
'''Note: This control does not affect PUCCH or PRACH.'''
total PUSCH power = p0_NominalPUSCH + 7.8 dBm
+
  
'''Note: This control does not affect PUCCH or PRACH, and so a raw power measurement of the UE transmitter may still show a different power level.'''
+
'''Note: This mode severely limits UL data throughput.'''
  
=== Getting PUSCH RSSI ===
+
=== Getting PUSCH power and path loss measurements ===
The ENB calculates PUSCH RSSI in dBm. These estimates are accurate to about +/-3 dbm. These measurements are for PUSCH only and do not include PUCCH or PRACH.
+
For PUSCH, the ENB measures:
 +
* power headroom, in dB per resource block
 +
* RSSI in dBm and dBm per resource block
 +
* path loss in dB
  
==== "enb uepool chans" ====
+
These measurements are accurate to about +/-3 db.
The [[LTE UE Connection Status|"enb uepool chans"]] command shows short-term average RSSI for each active UE under the heading "RSSI dBm".
+
These measurements are for PUSCH only and do not include PUCCH or PRACH.
  
Example with RSSI -32 dBm:
+
==== PHR in dB/RB? ====
<pre>
+
Q: What is power headroom in dB per resource block?
RNTI    PHR     RSSIdB  RSSIdBm UL-SNR  UL-FER  UL-HFER UL-eff  64QAM  ULbytes DL-eff  256QAM  DLbytes DL-NACK UC-RSSI UC-SNR  UC-miss Age    Idle    RSRP    RSRQ    UL-BSR  DL-BSR
+
61      30      -12    -32    20      2%      2%      15      false  2962    21      false  5314    0%      -26    43      0%      5      0      -100    -3      0      0
+
</pre>
+
  
==== "enb measurements pat= TB.UlRssiAvg" ====
+
A: It is the reported power headroom adjusted by the number of resource blocks in the UL assignment that carried the power headroom report.
The [[YateeNB KPI-related measurements]] track long-term average RSSI for all connected UEs.
+
The measurement name is "TB.UlRssiAvg".
+
<pre>
+
enb measurements pat= TB.UlRssiAvg
+
TB.UlRssiAvg=-56
+
LastClearTime=1529775690
+
CurrentTime=1529776604
+
</pre>
+
  
The "enb measurements clear" command clears and restarts the averaging period.
+
Specifically:
 +
PhrPerRb = PHRdB + 10 * log10(RBs)
  
=== UL Power Headroom ===
+
On a linear scale, this is the maximum number of resource blocks that can be sent on PUSCH, given the current UE PUSCH power control state.
The UE reports to the ENB a "power headroom" or "PHR", which is the number of dB between the UE's current transmission power and its maximum available transmission power.
+
For example, if a received PHR value is 10, that means that the PUSCH transport block that carried the PHR report was transmitted at 10 dB below the maximum power of the UE.
+
  
The ENB saves and expresses PHR values in dB-per-resource-block.
 
For example, if the ENB shows a PHR of 10 for a given UE, that means that a single-resource-block transmission would be 10 dB below the UE's maximum power, but a 10-resource-block transmission would be exactly at that maximum power.
 
(This information is used by the scheduler to avoid requesting more power than the UE can deliver.)
 
  
In fixed power mode, all PUSCH allocations are 6 resource blocks, so the actual transmit power of the UE is
+
==== Getting the information with "enb uepool chans" ====
UE transmit power = maximum power - PHR - 7.8 dBm
+
The [[LTE UE Connection Status|"enb uepool chans"]] command shows:
 +
* short-term average RSSI ''per resource block'' in dBm ("RSSIdBm")
 +
* power headroom ''per resource block'' in dB ("PHRdB")
 +
* path loss in dB ("Lp dB")
 +
for each active UE.
  
==== "enb uepool chans" ====
+
Example with two active UEs:
The "enb uepool chans" command shows short-term average PHR for each active UE under the heading "PHR".
+
 
+
Example with PHR of 30 dB:
+
 
<pre>
 
<pre>
RNTI    PHR    RSSIdB  RSSIdBm UL-SNR  UL-FER  UL-HFER UL-eff  64QAM  ULbytes DL-eff  256QAM  DLbytes DL-NACK UC-RSSI UC-SNR  UC-miss Age    Idle    RSRP    RSRQ    UL-BSR  DL-BSR
+
enb uepool chans
61     30     -12     -32     20     2%      2%      15      false  2962   21      false  5314    0%      -26     43     0%      5       0      -100   -3      0      0
+
RNTI    PHRdB  Lp dB  RSSIdB  RSSIdBm UL-SNR  UL-FER  UL-HFER UL-eff  64QAM  ULbytes DL-eff  256QAM  DLbytes DL-NACK UC-RSSI UC-SNR  UC-miss Age    Idle    RSRPdBm RSRQdB  UL-BSR  DL-BSR
 +
62     47      43     -25     -79     11     0%      4%      15      false  679k   153    false  11M    0%      -44     19     43%     121    0      -74    -3     0       0
 +
61      37      54      -24    -79    9       18%    0%      15      false  1555k  23      false  834k    0%      -28    39      1%      6912   0      -81    -3      0      0
 
</pre>
 
</pre>
  
==== "enb measurements pat= TB.UlRssiAvg" ====
+
With a fixed allocation size of 4 resource blocks, the total RSSI on PUSCH is 4 dB above the value reported here.
The YateeNB KPI-related measurements track long-term average PHR for all connected UEs.
+
In this example, both UEs have a total RSSI of -75 dBm.
The measurement name is "TB.UlPhrAvg".
+
 
 +
==== Getting the information with "enb measurements pat=" ====
 +
The [[YateeNB KPI-related measurements]] track these values for all connected UEs:
 +
* TB.UlPhrAvg - Average power headroom in dB ''per resource block''
 +
* TB.UlPathLossAvg - Average path loss in dB
 +
* TB.UlRssiAvg - Average RSSI in dBm
 +
* TB.UlRssiPerRbAvg" - Average RSSI ''per resource block'' in dBm
 +
 
 +
Example of accessing a measurement:
 
<pre>
 
<pre>
enb measurements pat= TB.UlPhrAvg
+
enb measurements pat= TB.UlRssiAvg
TB.UlPhrAvg=15
+
TB.UlRssiAvg=-56
LastClearTime=1529780687
+
LastClearTime=1529775690
CurrentTime=1529782244
+
CurrentTime=1529776604
 
</pre>
 
</pre>
  
 
The "enb measurements clear" command clears and restarts the averaging period.
 
The "enb measurements clear" command clears and restarts the averaging period.
  
== DL Power ==
+
 
 +
== DL power-related measurements ==
 
The UE reports to the ENB a received power level called "RSRP".  
 
The UE reports to the ENB a received power level called "RSRP".  
 
RSRP is the power level in dBm-per-resource element for the CSRS part of the LTE signal.
 
RSRP is the power level in dBm-per-resource element for the CSRS part of the LTE signal.

Revision as of 12:14, 18 July 2018

This page describes UL and DL power and path loss measurements that can be made with the YateENB, as of r7472.

UL power-related measurements

The ENB can provide real-time measurements of UL transmitted power, received power, and path loss.

  • Transmitted power is calculated from http://www.sharetechnote.com/html/Handbook_LTE_PHR.html reported power headroom], assuming that the UE is in the standard power class, with a maximum output of 23 dBm.
  • RSSI is calculated based on the known receiver gain for the radio hardware.
  • Path loss is calculated from transmitter power and RSSI. This is a port-to-port path loss, where all antennas and cables are included in the measured "path".

Setting fixed UL power on PUSCH

The ENB uses dynamic power control to adjust the output power of the UE so that the per-resource-block RSSI at the ENB receiver meets a configured target level. Because the number of resource blocks in an UL transmission vary depending on load, this means that the total transmitted power of the UE changes always even though the power per resource block is fixed.

To make path loss and power measurements easier from outside the ENB, the ENB supports fixed UL assignment sizes as a special test mode. This mode does not really fix the UL power to a constant level, but if the path loss is constant, UL power on PUSCH will also be constant.

To enable fixed UL assignment sizes, set the PuschFixedPower configuration parameter to true:

[basic]
; Fixed UL grant sizes for easier power measurement.
PuschFixedPower=true

This parameter is not supported in the LMI, and needs to be set in the enb-custom.conf file in the /etc/yate/sdr directory. (If you do not already have such a file, just create one. Be sure to indicate the "[basic]" section at the start of the file.)

The default size for the fixed UL grant is 4 resource blocks.

Note: This mode is not necessary for UL power, RSSI and power loss measurements.

Note: This control does not affect PUCCH or PRACH.

Note: This mode severely limits UL data throughput.

Getting PUSCH power and path loss measurements

For PUSCH, the ENB measures:

  • power headroom, in dB per resource block
  • RSSI in dBm and dBm per resource block
  • path loss in dB

These measurements are accurate to about +/-3 db. These measurements are for PUSCH only and do not include PUCCH or PRACH.

PHR in dB/RB?

Q: What is power headroom in dB per resource block?

A: It is the reported power headroom adjusted by the number of resource blocks in the UL assignment that carried the power headroom report.

Specifically:

PhrPerRb = PHRdB + 10 * log10(RBs)

On a linear scale, this is the maximum number of resource blocks that can be sent on PUSCH, given the current UE PUSCH power control state.


Getting the information with "enb uepool chans"

The "enb uepool chans" command shows:

  • short-term average RSSI per resource block in dBm ("RSSIdBm")
  • power headroom per resource block in dB ("PHRdB")
  • path loss in dB ("Lp dB")

for each active UE.

Example with two active UEs:

enb uepool chans
RNTI    PHRdB   Lp dB   RSSIdB  RSSIdBm UL-SNR  UL-FER  UL-HFER UL-eff  64QAM   ULbytes DL-eff  256QAM  DLbytes DL-NACK UC-RSSI UC-SNR  UC-miss Age     Idle    RSRPdBm RSRQdB  UL-BSR  DL-BSR
62      47      43      -25     -79     11      0%      4%      15      false   679k    153     false   11M     0%      -44     19      43%     121     0       -74     -3      0       0
61      37      54      -24     -79     9       18%     0%      15      false   1555k   23      false   834k    0%      -28     39      1%      6912    0       -81     -3      0       0

With a fixed allocation size of 4 resource blocks, the total RSSI on PUSCH is 4 dB above the value reported here. In this example, both UEs have a total RSSI of -75 dBm.

Getting the information with "enb measurements pat="

The YateeNB KPI-related measurements track these values for all connected UEs:

  • TB.UlPhrAvg - Average power headroom in dB per resource block
  • TB.UlPathLossAvg - Average path loss in dB
  • TB.UlRssiAvg - Average RSSI in dBm
  • TB.UlRssiPerRbAvg" - Average RSSI per resource block in dBm

Example of accessing a measurement:

enb measurements pat= TB.UlRssiAvg
TB.UlRssiAvg=-56
LastClearTime=1529775690
CurrentTime=1529776604

The "enb measurements clear" command clears and restarts the averaging period.


DL power-related measurements

The UE reports to the ENB a received power level called "RSRP". RSRP is the power level in dBm-per-resource element for the CSRS part of the LTE signal.

Total LTE DL power varies with activity. The total output power of and idle ENB, averaged over a frame, is about 10 dB higher than RSRP, depending on the bandwidth, CFI setting, and SIB sizes and scheduling. For an active ENB, the output is 20 - 30 dB higher than RSRP, depending on the configured bandwidth.

Getting RSRP

"enb uepool chans"

The "enb uepool chans" command shows short-term average RSRP for each active UE under the heading "RSRP".

"enb measurements pat= TB.UlRssiAvg"

The YateeNB KPI-related measurements track long-term average PHR for all connected UEs. The measurement name is "RRC.MeasReportRSRP".

enb measurements pat= RRC.MeasReportRSRP
RRC.MeasReportRSRP=-85
LastClearTime=1529780687
CurrentTime=1529782819

The "enb measurements clear" command clears and restarts the averaging period.