03 RA41333EN160GLA0 LTE Capacity Areas

RRM - Capacity Areas and Cell Resource Measurements Slide 1

NokiaEDU LTE Counters and KPI [FL16] RA4133 FL16 RRM - Capacity areas and cell resource measurements Module 3

RA41333EN160GLA0

© Nokia Solutions and Networks 2016

RA41333EN160GLA0

1


Copyright and confidentiality The contents of this document are proprietary and confidential property of Nokia Solutions and Networks. This document is provided subject to confidentiality obligations of the applicable agreement(s). This document is intended for use of Nokia Solutions and Networks customers and collaborators only for the purpose for which this document is submitted by Nokia Solutions and Networks. No part of this document may be reproduced or made available to the public or to any third party in any form or means without the prior written permission of Nokia Solutions and Networks. This document is to be used by properly trained professional personnel. Any use of the contents in this document is limited strictly to the use(s) specifically created in the applicable agreement(s) under which the document is submitted. The user of this document may voluntarily provide suggestions, comments or other feedback to Nokia Solutions and Networks in respect of the contents of this document ("Feedback"). Such Feedback may be

2

RA41333EN160GLA0

used in Nokia Solutions and Networks products and related specifications or other documentation. Accordingly, if the user of this document gives Nokia Solutions and Networks feedback on the contents of this document, Nokia Solutions and Networks may freely use, disclose, reproduce, license, distribute and otherwise commercialize the feedback in any Nokia Solutions and Networks product, technology, service, specification or other documentation. Nokia Solutions and Networks operates a policy of ongoing development. Nokia Solutions and Networks reserves the right to make changes and improvements to any of the products and/or services described in this document or withdraw this document at any time without prior notice. The contents of this document are provided "as is". Except as required by applicable law, no warranties of any kind, either express or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular

purpose, are made in relation to the accuracy, reliability or contents of this document. NOKIA SOLUTIONS AND NETWORKS SHALL NOT BE RESPONSIBLE IN ANY EVENT FOR ERRORS IN THIS DOCUMENT or for any loss of data or income or any special, incidental, consequential, indirect or direct damages howsoever caused, that might arise from the use of this document or any contents of this document. This document and the product(s) it describes are protected by copyright according to the applicable laws. Nokia is a registered trademark of Nokia Corporation. Other product and company names mentioned herein may be trademarks or trade names of their respective owners. © Nokia Solutions and Networks 2016


RA41333EN160GLA0

2


Course Objectives

Capacity areas and cell resource measurements After completing this learning element, the participant will be able to:

- Describe the different capacity areas in E-UTRAN - Explain outer and inner loop link quality control and the corresponding measurements - Aanalyze measurements regarding link adaptation, the selection of MCSs and adaptive transmission bandwidth - Explain the counters and KPIs related to packet scheduling, MIMO and power control

- Explain the counters and KPI’s for throughput and cell availability

4

RA41333EN160GLA0


RA41333EN160GLA0

3


Capacity areas and cell resource measurements • Capacity areas • LTE QoS concept • Outer and Inner Loop Link Quality Control • Inner Loop and Outer Loop Link Adaptation DL • Inner Loop and Outer Loop Link Adaptation UL • Adaptive Transmission Bandwidth • UL and DL scheduling • MIMO

• Power control • Cell throughput • Cell availability • Cell resource Groups

5

RA41333EN160GLA0


RA41333EN160GLA0

4


Capacity areas Measurement areas within NW architecture HSS

M8018 eNodeB Load Measurements

Number of UE

eNB Mobility Management Entity

See chapter 6 M8001 Cell Load Measurements M8011 Cell Resource Measurement S6a

M8012 Cell Throughput Measurements M8020 Cell Availability Measurements

MME

X2

M8005 UL Power and Quality Measurements M8010 DL Power and Quality Measurements

S1-U LTE-Uu Evolved Node B (eNB)

Serving Gateway

LTE-UE

6

M8004 Transport Measurements

M8013 UE state measurements

Data volume on X2

S1 setup

Throughput on X2

Number of S1 connections

RA41333EN160GLA0


RA41333EN160GLA0

5


Capacity areas Resource blocks & physical CH allocation (DL)

12 subcarriers = 180 kHz

72 central subcarriers = 6 RBs (minimum LTE Bandwidth = 1.4 MHz)

5 RBs allocated to UEa

1 RB allocated to UEb

PCFICH

PHICH

PDSCH UE1

DL RS

PDCCH

PDSCH UE2

Resource Block (RB)

7 OFDM symbols* x 12 subcarriers

Resource Element

12 subcarriers ..

.. Frequency Resource block

1 ms subframe or TTI

M8011 Cell Resource Measurements Allocation of physical resource blocks (DL) Utilization of physical resource blocks (DL)

0.5 ms slot Time

7

RA41333EN160GLA0


RA41333EN160GLA0

6


Capacity areas Resource blocks & physical CH allocation (UL) PUCCH & PUSCH allocation Physical Uplink Control Channel Subframe = 1 ms

1 Resource Block RB carries UL control information (in the absence of UL data) PUCCH

total UL bandwidth

The PUCCH is never transmitted simultaneously with the PUSCH from the same UE

frequency

Frequency hopping

   

PUSCH

PUCCH M8011 Cell Resource Measurements Allocation of physical resource blocks (UL)

Slot

Utilization of physical resource blocks (UL)

8

RA41333EN160GLA0


RA41333EN160GLA0

7


TDD Frame Structure • TDD utilizes a Type 2 frame structure. – Frame, subframe and slot duration (except special slots) are the same as FDD (Type 1). – It enables a static configuration between downlink and uplink subframe. • Each frame carries 10 subframes and depending on frame configuration one or two Special Subframes. • Special Subframes utilize three specialized field: DwPTS, GP, UpPTS.

SF #4

SF #5

UpPTS

SF #3

GP

SF #2

DwPTS

UpPTS

SF #0

GP

Radio Frame 10ms DwPTS

f UL/DL carrier

SF #7

SF #8

SF #9

Subframe 1ms Downlink Subframe DwPTS: Downlink Pilot time Slot Uplink Subframe

time

UpPTS: Uplink Pilot Time Slot

Special Subframe

GP: Guard Period to separate DL or UL Subframe between DL/UL

9

RA41333EN160GLA0


RA41333EN160GLA0

8


TDD Frame Configuration (Subframe Assignment) -

There are 7 frame configurations (sa0, sa1, sa2, sa3, sa4, sa5, sa6), enabling different DL/UL ratios. tddFrameConf TDD Uplink-downlink Configuration parameter LNCEL; 1....2; - ; 1

1 frame = 10ms 1 subframe = 1ms

0

DL

S

UL

UL

UL

DL

S

UL

UL

UL

1

DL

S

UL

UL

DL

DL

S

UL

UL

DL

2

DL

S

UL

DL

DL

DL

S

UL

DL

DL

3

DL

S

UL

UL

UL

DL

DL

DL

DL

DL

4

DL

S

UL

UL

DL

DL

DL

DL

DL

DL

5

DL

S

UL

DL

DL

DL

DL

DL

DL

DL

6

DL

S

UL

UL

UL

DL

S

UL

UL

DL

10

RA41333EN160GLA0


RA41333EN160GLA0

9


Capacity Related Measurements (Overview) • M8001 Cell Load Measurements – PDCP SDU delay on DL/UL DTCH – RACH setup – Transmitted Transport blocks on PCH, BCH and SCH (UL/DL) – Transmission on PDSCH and PUSCH classified per MCS – The number of transmissions on PDSCH over the measurement period using MCS – The number of unsuccessful transmissions on PDSCH using MCS – The number of unacknowledged transmissions on PDSCH using MCS – RLC SDUs on PCCH, BCCH, CCCH (UL/DL), DCCH (UL/DL) and DTCH (UL/DL) – PDCP SDUs on DTCH (UL/DL) – Number of Ues – Average Dl latency per GBR QCI – Overload duration – BLER based on NACK/ACK ratio – Carrier aggregation UEs • M8011 Cell Resource Measurements – Allocation of physical resource blocks (UL/DL) per QoS and SRB – Utilization of physical resource blocks (UL/DL) per QoS and SRB – PDCCH usage per Aggregation level – Number of OFDM symbols used for subframe signaling – Composite available capacity – eICIC – CA PUCCH blocking

11

RA41333EN160GLA0


RA41333EN160GLA0

10


M8001 Cell Load Measurements Examples

PDSCH Usage per MCS over 1 hour

Mean delay 70 60

10,00,000 1,00,000

50

10,000

40

1,000

PDCP_RET_DL_D EL_MEAN_QCI_1

30

100

PDCP_RET_DL_D EL_MEAN_NON_ GBR

20

10

10

1

00:00:00 02:00:00 04:00:00 06:00:00 08:00:00 10:00:00 12:00:00 14:00:00 16:00:00 18:00:00 20:00:00 22:00:00

0

12

RA41333EN160GLA0


RA41333EN160GLA0

11


M8011 Cell Resource Measurements examples

DL PRB Utilization in 1 hour 100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

Aggregation group PDCCH utilization 25,00,00,000 20,00,00,000 15,00,00,000 10,00,00,000 5,00,00,000 0 AGG1, used AGG2, used AGG4, used AGG8, used

13

RA41333EN160GLA0


RA41333EN160GLA0

12

RRM - Capacity Areas and Cell Resource Measurements

overload level 0 (normal operation) U-plane overload level 0 (UPOVL 0) represents the normal operation. There is no U-plane overload detected, no U-plane overload countermeasures are active and the KPIs are no impacted due to U-plane overload (assuming no other overload e.g. C-plane overload is active at this time).

overload level 1 (for graceful overload handling) Within this overload level the U-plane overload countermeasures for UPOVL 1 will be started (if activated). The goal of these countermeasures is to avoid further U-plane traffic increase and to graceful reduce U-plane load. At his level, ongoing services shall be maintained and emergency and high priority traffic shall be prioritized. Other activated features might be slightly impacted.

overload level 2 (preserving stability, self-defense U-plane overload handling) Within this overload level the U-plane overload countermeasures UPOVL 2 will be started (if activated) or upgraded in case they were already started for UPOVL1. The goal of these countermeasures is not only to avoid further U-plane traffic increase but to reduce U-plane load immediately also in a non-graceful manner. At this level, ongoing services might be impacted and only emergency or high priority traffic shall be admitted. In addition other features might be heavily impacted and/or switched off completely.

RA41333EN160GLA0

14


LTE1804: Downlink carrier aggregation 3CC - 60 MHz

3CC 60 MHz CA principles: • Enables aggregation of three CC • Contiguous and non-contiguous intra-band and inter-band CA supported

band 1

Intra-band, contiguous

CA capable UE

f band 1 CA noncapable UE

Intra-band, noncontiguous f band 1

band 2

Inter-band, non-contiguous f

14

RA41333EN160GLA0

M8001C494 Average number of DL carrier aggregated capable UEs M8001C495 Average number of UEs with a configured SCell M8001C496 Average number of UEs with an activated SCell


RA41333EN160GLA0

13


LTE1804: Downlink carrier aggregation 3CC - 60 MHz UE level – configuration of SCell(s) • Configuring SCell = adding SCell • Active UE – UE in RRC Connected state with DRB established (with or without data in the buffer)

• Up to 100 active UEs can use CA: − up to 100 of them can be configured with a SCell − up to 100 of them, can be configured with two SCells

CA UEs with SCell configured

CA UEs with two SCells configured

100 Maximum number of active UEs

Cell configuration limitations:

CA active UEs

50 20

LNCELmaxNumCaConfUe

LNCELmaxNumCaConfUeDc

LNCELmaxNumCaConfUe3c

• Until the conditions satisfied: max (

maxNumCaConfUeDc

,

maxNumCaConfUe3c

)≤

maxNumCaConfUe

sum (

maxNumCaConfUeDc

,

maxNumCaConfUe3c

)≥

maxNumCaConfUe

15

RA41333EN160GLA0


RA41333EN160GLA0

14


Carrier Aggregation

eNB

UE

M8011C67 Number of SCell configuration attempts

RRC:RRCConnectionReconfiguration (SCellToAddModList-r10)

M8011C68 Number of successful SCell configurations RRC:RRCConnectionReconfigurationComplete (SCellToAddModList-r10)

Cell A

Cell B

PDCP

PDCP

RLC

RLC

MAC

MAC

PHY

PHY

RLC_PDU_VOL_TRANSMITTED

RLC_PDU_DL_VOL_CA_SCELL

M8012C151

PCell RLC data volume in DL via SCell

PCell A Scell for Pcell B

16

RA41333EN160GLA0

Pcell B Scell for Pcell A


RA41333EN160GLA0

15


CA PUCCH blocking • If CA is activated for a UE, there will be a need to transmit multi-cell ACK/NACK messages – reports covering simultaneously all transport blocks for all cells (PCell and up to two SCells) • With only 1 Scell “Format 1b With Channel Selection” can be used, ( this alternates the ack nack messages alternatively for TB1 then TB2).

TB1 PCell

TB2 PCell

TB1 SCell 1

TB2 SCell 1

TB1 SCell 2

TB2 SCell 2

ACK/NACK

ACK/NACK

ACK/NACK

ACK/NACK

ACK/NACK

ACK/NACK

• With 2 Scells more data needs to be sent in a single ACK/NACK report than normally available so PUCCH Format 3 is used

17

RA41333EN160GLA0

M8011C166

SCell scheduling blocking rate due to conflicts on PUCCH format 1bwcs resources

M8011C167

SCell scheduling blocking rate due to conflicts on PUCCH format 3 resources © Nokia Solutions and Networks 2016

PUCCH format 3 introduced in release 10 (designed specifically for Carrier Aggregation deployments with more than one SCell, where more than 4 bits of HARQ information have to be reported)

RA41333EN160GLA0

16


Downlink carrier aggregation Performance management Counter number

Counter name

Description

M8001C494

CA_DL_CAP_UE_AVG

Average number of DL carrier aggregated capable UEs

M8001C495

CA_SCELL_CONF_UE_AVG

Average number of UEs with a configured Scell

M8001C496

CA_SCELL_ACTIVE_UE_AVG

Average number of UEs with an activated Scell

M8011C67

CA_SCELL_CONFIG_ATT

Number of SCell configuration attempts

M8011C68

CA_SCELL_CONFIG_SUCC

Number of successful SCell configurations

M8012C151

RLC_PDU_DL_VOL_CA_SCELL

PCell RLC data volume in DL via Scell

M8001C497

CA_DL_CAP_UE_3CC_AVG

Average number of DL carrier aggregated capable UEs for 3 CCs

M8001C498

CA_2SCELLS_CONF_UE_AVG

Average number of UEs with two configured SCells

M8001C499

CA_2SCELLS_ACTIVE_UE_AVG

Average number of UEs with two activated Scells

M8011C165

CA_SCELL_SWAP_A6

Number of Event A6 triggered SCell swaps

M8011C166

PUCCH_BLOCK_RATE_FORMT_1 BWCS

SCell scheduling blocking rate due to conflicts on PUCCH format 1bwcs resources

M8011C167

PUCCH_BLOCK_RATE_FORMT_3

SCell scheduling blocking rate due to conflicts on PUCCH format 3 resources

18

RA41333EN160GLA0


RA41333EN160GLA0

17


t0+N

t0

Serving cells max capacity

• Decision to add SCell is based on the data amount/activity monitoring (non-GBR DL data buffer)

time

New triggers added for SCell deactivation and deconfiguration • Decision to release SCell is based on the reported Channel Quality Indicator for this SCell

New counters: M8011C168 CA_SCELL_SWAP_BLIND_A6 M8011C169 CA_SCELL_SWAP_BLIND_NA6

CQI <= T1 SCell deconfiguration

t0

t0+N

CA_SCELL_SWAP_BLIND_NA6 (LTE_5902b) =

Blind SCell Configuration with A6 Distribution Ratio (LTE_5903b) =

RA41333EN160GLA0

T1 < CQI < T2 SCell deactivation

time

New KPIs:

Blind SCell Configuration without A6 Distribution Ratio

19

SCell addition after ’N’ measurement periods

Non-GBR DL buffer

• New conditional approach for SCells addition and SCells release

CQI

LTE1541 Specific functionality

FDD/TD-LTE 16 basic functionality

LTE1541 Advanced SCell measurement handling

×100%

CA_SCELL_CONFIG_SUCC + CA_SCELL_SWAP_A6 + CA_SCELL_SWAP_BLIND_A6+ CA_SCELL_SWAP_BLIND_NA6

CA_SCELL_SWAP_BLIND_A6

×100%

CA_SCELL_CONFIG_SUCC + CA_SCELL_SWAP_A6 + CA_SCELL_SWAP_BLIND_A6+ CA_SCELL_SWAP_BLIND_NA6


RA41333EN160GLA0

18


Capacity Related Measurements (Overview) •M8012 Cell Throughput Measurements – Data volume on BCH and SCH (UL/DL) – MAC-PDU volume on PUSCH and PDSCH – MAC-SDU volume on PCCH, BCCH, CCCH (UL/DL), DCCH (UL/DL), DTCH (UL/DL)

– RLC-SDU volume and throughput on DCCH (UL/DL) and DTCH (UL/DL)

– PDCP-SDU volume and throughput (UL/DL) – IP throughput UL/DL per QCI •M8020 Cell Availability Measurements – Cell availability and unavailability •M8005 UL Power and Quality Measurements – RSSI on PUCCH and PUSCH – UE power control headroom on PUSCH – SINR on PUCCH and PUSCH – UL upgrade / downgrade by AMC – Total received Power – UL Interference – SIR per PRB measurements

20

RA41333EN160GLA0

•M8010 DL Power and Quality Measurements – CQI- UE Reported CQI Level 00, e.g. CQI 0 is reported by UE. (From CQI 0 to 15) CQI offset MIMO modes PDCCH used for retransmission AVG and Max Tx power Beamforming modes (TDD only) CQI for codeword 1 • M8031 SINR Measurement • Wideband and narrow band DL SINR

– – – – – –


RA41333EN160GLA0

19


M8012 Cell Throughput Measurements Examples

21

RA41333EN160GLA0

23:00:00

21:00:00

19:00:00

17:00:00

15:00:00

13:00:00

11:00:00

PDCP SDU, DL

03:00:00

DL-DTCH

PDCP SDU, UL

09:00:00

UL-DTCH

8,00,000 7,00,000 6,00,000 5,00,000 4,00,000 3,00,000 2,00,000 1,00,000 0 07:00:00

800.00 700.00 600.00 500.00 400.00 300.00 200.00 100.00 0.00

Total PDCP SDU volume (kB)

05:00:00

MAC SDU volume per channel (MB)


RA41333EN160GLA0

20


M8020 Cell Availability Measurements Example

Cell Availabilty 100.50%

1.80% 1.60%

100.00%

1.40% 99.50%

1.20% 1.00%

99.00% 0.80%

Cell Availability Cell Planned Unavailabilty

98.50%

0.60%

Cell Unplanned availability

0.40% 98.00% 0.20% 97.50%

22

0.00%

RA41333EN160GLA0


RA41333EN160GLA0

21


M8005 UL Power and Quality Measurements Examples

Power headroom reports

RSSI values

1,20,000

-90.00 -92.00 -94.00 -96.00 -98.00 -100.00 -102.00 -104.00 -106.00 -108.00 -110.00

23

1,00,000 80,000 60,000 RSSI_PUCCH_AVG

40,000

RSSI_PUSCH_AVG

20,000 0

RA41333EN160GLA0


RA41333EN160GLA0

22


M8005 DL Power and Quality Measurements Examples 24Hr CQI distribution LTE_5427A Average CQI 14.6 14.4 14.2 14.0 13.8 13.6 13.4 13.2 13.0 12.8

Level 00 Level 01 Level 02 Level 03 Level 04 Level 05 LTE_542…

Level 06 Level 07 Level 08 Level 09 Level 10 Level 11 Level 12

Level 13 Level 14 Level 15

24

RA41333EN160GLA0


RA41333EN160GLA0

23


M8010 DL Power and Quality Measurements Examples 70.00

900 800

60.00

700 50.00

600

40.00

500

30.00

400 300

20.00

OL div OL spatial mux MIMO mode switches

200 10.00

100 0 11:00:00 13:00:00 15:00:00 17:00:00 19:00:00 21:00:00 23:00:00 01:00:00 03:00:00 05:00:00 07:00:00 09:00:00

0.00

100.00

8,000

90.00

7,000

80.00

6,000

70.00 60.00

5,000

50.00

4,000

CL, Single Codeword

40.00

3,000

CL, Double Codeword

30.00

2,000

20.00

MIMO mode switches

1,000

10.00

0 12:00:00 13:00:00 14:00:00 15:00:00 16:00:00 17:00:00 18:00:00 19:00:00 20:00:00 21:00:00 22:00:00 23:00:00 00:00:00

0.00

25

RA41333EN160GLA0


RA41333EN160GLA0

24

26 UE_REP_CQI_Level 0

…………..

RA41333EN160GLA0

RA41333EN160GLA0 UE_REP_CQI_CW1_Level 3

UE_REP_CQI_CW1_Level 2

…………..




Wideband CQI reports on CW0



UE_REP_CQI_Level 15

UE_REP_CQI_Level 14

UE_REP_CQI_Level 13

UE_REP_CQI_Level 3

UE_REP_CQI_Level 2

UE_REP_CQI_Level 1


CQI reporting

• From FL15a onwards: includes reporting on the CQI of CW 1 Wideband CQI reports on CW1


25


Capacity areas and cell resource measurements • Capacity areas • LTE QoS concept • Outer and Inner Loop Link Quality Control • Inner Loop and Outer Loop Link Adaptation DL • Inner Loop and Outer Loop Link Adaptation UL • Adaptive Transmission Bandwidth • UL and DL scheduling • MIMO • Power control • Cell throughput • Cell availability • Cell resource Groups

27

RA41333EN160GLA0


RA41333EN160GLA0

26


QoS Functions QoS functions have not been standardized in detail in 3GPP but are implementation specific • QoS functions • Admission control (RL40 → Smart admission control) • Bandwidth management based on UE-AMBR (AMBR = aggregate maximum bit rate)

• Bearer and service level bandwidth management • Bearer and service level DSCP marking (DSCP = differential service code point) • Queuing and scheduling of packets • Allocation and Retention Policy (ARP)

28

RA41333EN160GLA0


RA41333EN160GLA0

27


QoS Parameters (EPS) • QoS Class Identifier (QCI) •

Used to determine packet forwarding treatment (e.g. scheduling of packets)

•

Used to mark packets with DSCP (Differential Service Code Point)

•

3GPP has standardized 9 QCI values and mapped to resource type (GBR, non-GBR), priority, packet delay budget and packet error loss rate

• Allocation and Retention Priority (ARP) •

Used to decide whether bearer establishment or modification request can be accepted in case of resource limitations

•

Can also be used to decide which bearer to drop during resource limitations

•

According 3GPP ARP has no impact on packet forwarding treatment

• APN AMBR and UE AMPR for non-GBR EPS bearers •

APN-AMBR shared by all non-GBR EPS bearers with the same APN – downlink enforcement is done in PDN GW and uplink enforcement in UE

•

UE-AMBR shared by all non-GBR EPS bearers of the UE – downlink and uplink enforcement is done in eNB

• Guaranteed Bit Rate (GBR) and Max Bit Rate (MBR) for GBR EPS bearers • Nominal Bit rate for nGBR bearers - grant the non-GBR bearers (with the NBR defined), additional resources (PRBs) to meet the defined nominal bit rate

29

RA41333EN160GLA0


RA41333EN160GLA0

28


LTE1042: Nominal Bitrate for non-GBR bearers NBR, plain non-GBR and GBR bearers bitrate

bitrate

Guaranteed bitrate

Non-guaranteed bitrate

GBR value

time

time bitrate

Nominal bitrate Commited value

NBR value Achieved value

M8006C233

ERAB_NBR_DL_AVG

M8006C234

ERAB_NBR_UL_AVG

M8006C235

ERAB_NBR_DL_FAIL_OVL_AVG

M8006C236

ERAB_NBR_UL_FAIL_OVL_AVG

time

30

RA41333EN160GLA0


Non-GBR non-NBR bearers („legacy” non-GBR): •Always served on best effort basis •No committments as to bitrate or delay (only minimal bitrate is guaranteed) GBR bearers: •The commited bitrate is guaranteed •The commited bitrate can be exceeded •Delay control •If the commited bitrate can not be met, GBR bearers will be torn down •Aggregated GBR capacity is considered by admission control Non-GBR NBR bearers: •The commited bitrate is not guaranteed •The commited bitrate can be exceeded •No delay control •bearers will not be torn down when the declared bitrate cannot be met •NBR capacity is not considered at admission control

RA41333EN160GLA0

29


Service Differentiation for Non-GBR EPS Bearer QCI (QoS Class Identifier) based service differentiation • Support of QCI 1,2,3,4 GBR QCI Classes with • •

•

Delay based scheduling and priority classification. Support of up to 21 Operator Specific QCI’s in the range 128 to 254 non GBR Differentiation of 5 different non-GBR QCI classes with relative scheduling weights -QCIs: 5,6,7,8,9 Flexi Multiradio allows to assign relative scheduling weights for each non GBR QCI on cell level The relative weight or delay is considered by the UL and DL scheduler

• Default bearers are set up with QCI 9 (for non-

QCI Resource Priority Packet Packet Type Delay Error Budget Loss 2

100 ms

Rate 10-2

2

4

150 ms

10-3

3

3

50 ms

10-3

4

5

300 ms

10-6

1

100 ms

10-6

6

300 ms

10-6

7

100 ms

10-3

8

300 ms

10-6

9

300 ms

10-6

1

Example Services

Conversational Voice

GBR

5 6

privileged users) or QCI 8 (for premium users)

Non-GBR

7

Conversational Video (Live Streaming) Real Time Gaming Non-Conversational Video (Buffered Streaming) IMS signaling Video (Buffered Streaming) TCP-based (e.g., www, e-mail, chat, ftp, p2p file sharing, progressive video, etc.) Voice, Video (Live Streaming) Interactive Gaming

8 Feature ID(s): LTE9, LTE10 ,LTE496, LTE518

9

Video (Buffered Streaming) TCP-based (e.g., www, e-mail, chat, ftp, p2p file sharing, progressive video, etc.)

31

RA41333EN160GLA0


RL30: Operator specific CQI (LTE418): 21 new QCIs configurable in the range from 128 to 254. Better user and service differentiation for non-GBR services, e.g. bronze, silver and gold users. Operator specific QCIs can be used as well in case of RAN sharing to define a set of QCIs dedicated for each operator

RA41333EN160GLA0

30


Layer 4 to layers 3&2 mapping - Services are transferred by bearers which are mapped to QCIs. - At the edge of the transport domain QCIs are mapped to DSCP codes and then associated with respective DiffServ PHB classes.

LTE Radio domain LTE Traffic Class

QCI

LTE Transport domain Resourc e Type

DiffSer v PHB

Ethernet p-bits

46

EF

7

36

AF42

5

46

EF

7

5

26

AF31

3

Voice, video, interactive gaming

6

34

AF41

4

18

AF21

2

Video (buffered streaming)

7

20

AF22

2

TCP-based (e.g. www, email, ftp, p2p filesharing, etc.)

8

10

AF11

1

9

0

BE

0

Conversational Voice

1

Conversational Video

2

Real-time Gaming

3

Non-conversational Video

4

IMS singaling

32

RA41333EN160GLA0

GBR

NonGBR

DSCP

Descriptions: DSCP Differentiale Service Code Point. Diff. Serv. (QoS) uses 6-bit DSCP field in the header of IP packets for packet classification purposes EF expedited forwarding (highest IP service class) AF assured forwarding (4 IP service classes, 4 / 1 = very high / low priority) BE best effort (lowest IP service class) IP QoS parameters packet delay / loss / jitter


RA41333EN160GLA0

31


Measurement support for QoS M8001C227, 228, 229, 230, 235

UEs with buffered DL data for DRB with QCI 1, 2, 3, 4 and non-GBR

M8001C269

Mean PDCP SDU delay on DL DTCH for DRB with QCI 1

M8001C270

Mean PDCP SDU delay on DL DTCH for non-GBR DRB

M8001C271, 273, 273

PDCP SDU delay on DL DTCH Mean for GBR DRBs of QCI 2, 3, 4

LTE_5471a, LTE_5472a, LTE_5473a, LTE_5474a, LTE_5475a, LTE_5476a, LTE_5477a, LTE_5478a, LTE_5479a LTE_5304b LTE_5305b LTE_5306b LTE_5307b LTE_5308b LTE_5450a LTE_5451a LTE_5452a LTE_5453a LTE_5454a LTE_5310b LTE_5311b LTE_5312b LTE_5313b LTE_5314b LTE_5455a LTE_5456a LTE_5457a LTE_5458a LTE_5459a

33

E-UTRAN Averaged PDCP SDU Delay in DL, QCI1……QCI9

E-UTRAN PDCP SDU Loss Ratio in the DL, QCI1…QCI9

E-UTRAN PDCP SDU Loss Ratio in the UL, QCI1…QCI9

M8001C305, 306, 307, 308

PDCP SDU UL QCI 1, 2, 3, 4

M8001C309, 310, 311, 312, 313

PDCP SDU delay on DL DTCH Mean for non-GBR DRBs of QCI 5, 6, 7, 8, 9

M8001C314, 315, 316, 317

PDCP SDU DL QCI 1, 2, 3, 4

M8001C323, 324, 325, 326

PDCP SDU discarded DL QCI 1, 2, 3, 4

M8006C176, 177, 178, 179

Released active ERABs QCI1,QCI2, QCI3, QCI4

M8006C238, 239,240, 241, 242,243

Released active ERABs QCI5, QCI6, QCI7, QCI8, QCI9

M8000C181, 182, 183, 184

In-session activity time for QCI1, QCI2, QCI3, QCI4 ERABs

RA41333EN160GLA0


RA41333EN160GLA0

32


Measurement support for QoS

34

M8006C233

Non-GBR E-RABs with configured NBR in DL

M8006C234

Non-GBR E-RABs with configured NBR in UL

M8006C235

Non-GBR E-RABs not reaching NBR in DL due to overload

M8006C236

Non-GBR E-RABs not reaching NBR in UL due to overload

M8006C188 - 196 M8006C197 - 205 M8006C206 - 214 M8006C215 - 223 M8006C224 - 232 M8006C233 M8006C234 M8006C235 M8006C236

Setup attempts for initial E-RABs of QCI1-9 Setup attempts for additional E-RABs of QCI1-9 Successfully established initial E-RABs of QCI1-9 Successfully established additional E-RABs of QCI1- 9 Maximum number of simultaneous E-RABs of QCI1-9 Non-GBR E-RABs with configured NBR in DL Non-GBR E-RABs with configured NBR in UL Non-GBR E-RABs not reaching NBR in UL due to overload Non-GBR E-RABs not reaching NBR in DL due to overload

RA41333EN160GLA0


RA41333EN160GLA0

33



35

RA41333EN160GLA0


RA41333EN160GLA0

34


Inner Loop Link Quality Control • Link quality must be known to decide about - Modulation (UL/DL) - Coding (UL/DL) - MIMO mode (DL) • The decision is taken on the basis of - CQI (DL) cell cell

- BLER (UL)

cell

CQI measurements (inner loop) CQI offset measurements (outer loop)

36

RA41333EN160GLA0

M8010C36…C51 M8010C52/C53/C54


RA41333EN160GLA0

35


Outer Loop Link Quality Control (DL) •

BLER measurements on UL give true picture of link quality

•

But CQI measurements for DL performed by UE may be affected by many errors

– CQI estimation error of the UE – CQI quantization error – CQI reporting error – Time delay between CQI measurement and the reception of the subsequent data block – CQI interpolation error – Errors due to CQI averaging of physical resource blocks •

CQI values reported by UE therefore corrected on the basis of DL BLER measurements (ACK/NACK feedback of UE)

– If BLER better than dlTargetBler (LNCEL, 0.1%..99.9%, 0.1%), better CQI can be adopted CQIcorr = CQImeas + CQIstepup

– If BLER worse than dlTargetBler, lower CQI has to be adopted CQIcorr = CQImeas – CQIstepdown

– The correction depends on the deviation of the actual BLER from the BLER target (hardcoded) •

37

The correction of the reported CQI is optional and must be enabled with dlOlqcEnable (LNCEL, false/true, true)

RA41333EN160GLA0


RA41333EN160GLA0

36


Outer Loop Link Quality Control (DL)

38

RA41333EN160GLA0


RA41333EN160GLA0

37


Quality Related Measurements DL • CQI measurements (inner loop) -

M8010C36..C51: CQI histogram

-

The histograms work as follows 1.st counter: Number of measurements with CQI = 0

-

2.nd counter: Number of measurements with CQI = 1 Continue with steps of 1

Last counter: Number of measurements with CQI = 15

•CQI offset measurements (outer loop) -

39

M8010C52/C53/C54: Min/Max/Mean CQI offset (given in units of 0.001)

RA41333EN160GLA0


RA41333EN160GLA0

38


Quality Related Measurements DL UE Reported CQI

• CQI measurements (inner loop) 450 400 350 300 250 200 150

UE Reported CQI

100 50 0

40

RA41333EN160GLA0


Live cell data taken 09.07.13,08:00

RA41333EN160GLA0

39


Quality Related KPIs DL LTE_5427a E-UTRAN Average CQI Shows the average UE reported Channel Quality Indicator (CQI) value

Summarization formula (PI ID)

Logical formula AVG CQI= sum (number of hits in class_x * x) / Sum (total number of hits over all classes) x = 0, ..., 15

Sum (1*[M8010C37] + 2*[M8010C38] + 3*[M8010C39] + 4*[M8010C40] + 5*[M8010C41] + 6*[M8010C42] + 7*[M8010C43] + 8*[M8010C44] + 9*[M8010C45] + 10*[M8010C46] + 11*[M8010C47] + 12*[M8010C48] + 13*[M8010C49] + 14*[M8010C50] + 15*[M8010C51]) / Sum ([M8010C36] + [M8010C37] + [M8010C38] + [M8010C39] + [M8010C40] + [M8010C41] + [M8010C42] + [M8010C43] + [M8010C44] + [M8010C45] + [M8010C46] + [M8010C47] + [M8010C48] + [M8010C49] + [M8010C50] + [M8010C51])

Note; CQI 0 not included in numerator only in denominator

41

RA41333EN160GLA0

Summarization formula (Abbreviation) Sum (1*[UE_REP_CQI_LEVEL_01] + 2*[UE_REP_CQI_LEVEL_02] + 3*[UE_REP_CQI_LEVEL_03] + 4*[UE_REP_CQI_LEVEL_04] + 5*[UE_REP_CQI_LEVEL_05] + 6*[UE_REP_CQI_LEVEL_06] + 7*[UE_REP_CQI_LEVEL_07] + 8*[UE_REP_CQI_LEVEL_08] + 9*[UE_REP_CQI_LEVEL_09] + 10*[UE_REP_CQI_LEVEL_10] + 11*[UE_REP_CQI_LEVEL_11] + 12*[UE_REP_CQI_LEVEL_12] + 13*[UE_REP_CQI_LEVEL_13] + 14*[UE_REP_CQI_LEVEL_14] + 15*[UE_REP_CQI_LEVEL_15]) / Sum ([UE_REP_CQI_LEVEL_00] + [UE_REP_CQI_LEVEL_01] + [UE_REP_CQI_LEVEL_02] + [UE_REP_CQI_LEVEL_03] + [UE_REP_CQI_LEVEL_04] + [UE_REP_CQI_LEVEL_05] + [UE_REP_CQI_LEVEL_06] + [UE_REP_CQI_LEVEL_07] + [UE_REP_CQI_LEVEL_08] + [UE_REP_CQI_LEVEL_09] + [UE_REP_CQI_LEVEL_10] + [UE_REP_CQI_LEVEL_11] + [UE_REP_CQI_LEVEL_12] + [UE_REP_CQI_LEVEL_13] + [UE_REP_CQI_LEVEL_14] + [UE_REP_CQI_LEVEL_15])


RA41333EN160GLA0

40


Quality Related KPIs DL

11.5

11.0

10.5

10.0

9.5

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

9.0

LTE_5427a Average CQI…

42

RA41333EN160GLA0


RA41333EN160GLA0

41


Quality Related KPIs DL LTE_5432b E-UTRAN Average CQI Offset - FDD Shows the average eNodeB used offset (correction) value for Channel Quality Indicators (CQI)

Logical formula AVG CQI Offset = average of measured CQI offset values

43


Summarization formula (Abbreviation)

Avg ([M8010C54])

Avg ([CQI_OFF_MEAN])

RA41333EN160GLA0


RA41333EN160GLA0

42



44

RA41333EN160GLA0


RA41333EN160GLA0

43


Inner Loop Link Adaptation (DL) Actual MCS will be performed by BLER/CQI. In case of retransmission the same MCS is taken as for the original

BLER target

Actual BLER

Link adaptation adjustment of CQI measurements will be performed by UE

BLER = f (reported CQI, (respectively SINR)) Actual CQI is estimated by UE including ACK/NACK feedback from L1/L2, reported to eNodeB

PwR

Orig i

MCS 20

MCS 16 5/5 CQI 15

nR etra

64QAM

nal

Tran s

ACK /

3/5 CQI 10

mis

NAC K

nsm

issi

ons

MCS 9 2/5

16QAM

1/5 QPSK

eNodeB

45

sion

Link adaptation

DL Adaptive Modulation & Coding shall select its scheme table according to the radio CH conditions (CQI) RA41333EN160GLA0


RA41333EN160GLA0

44


Link Adaptation for User Data (PDSCH) • Inner Loop - A certain MCS is selected on the basis of the CQI reported by the UE - In case of retransmission the same MCS is taken as for the original • Outer Loop: • - The reported CQI is corrected on the basis of BLER measurements (already discussed) • Each service starts with an initial MCS defined by the parameter iniMcsDl • The feature must be enabled with the parameter dlamcEnable

46

iniMcsDl

dlamcEnable

LNCEL, 0..28, 4

LNCEL, true/false, true

RA41333EN160GLA0


RA41333EN160GLA0

45


MCSs for DL DL MCSs

MCS

ITBS

0-QPSK 1-QPSK 2-QPSK 3-QPSK 4-QPSK 5-QPSK 6-QPSK 7-QPSK 8-QPSK 9-QPSK 10-16QAM 11-16QAM 12-16QAM 13-16QAM 14-16QAM 15-16QAM 16-16QAM 17-64QAM 18-64QAM 19-64QAM 20-64QAM 21-64QAM 22-64QAM 23-64QAM 24-64QAM 25-64QAM 26-64QAM 27-64QAM 28-64QAM

0 1 2 3 4 5 6 7 8 9 9 10 11 12 13 14 15 15 16 17 18 19 20 21 22 23 24 25 26

47

MCS_index Mod order 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 RA41333EN160GLA0

2 2 2 2 2 2 2 2 2 2 4 4 4 4 4 4 4 6 6 6 6 6 6 6 6 6 6 6 6

- DL AMC shall select the MCS to be employed from the tables according to the radio conditions - On DL all MCSs specified by 3GPP are available

64QAM 6 bits/symbol

16QAM QPSK

4 bits/symbol

2 bits/symbol 64QAM 16QAM

b0 b1 b2 b3

QPSK b0 b 1 Im 01 11

00

b0 b1 b2 b3 b4 b5 Im

Im 1111

Re

10Re

Re

0000


RA41333EN160GLA0

46


Link Adaptation Measurements for DL • MCS usage -

M8001C45..C73: Histogram for number of PDSCH transmissions with MCS0..MCS28

-

M8001C103..C131: Histogram for number of not acknowledged PDSCH transmissions with MCS0..MCS28

-

M8001C156..C176 and M8001C202..C209: Histogram for discarded PDSCH transmissions with MCS0..MCS20 and MCS21..MCS28 due to maximum number of retransmissions

M8001C45..C73 Total transmission

48

M8001C103..C131 Transmission with feedback NACK

RA41333EN160GLA0

M8001C156..176 and M8001C202..209 Discarded transmission


RA41333EN160GLA0

47

49

RA41333EN160GLA0

RA41333EN160GLA0 PDSCH_TRANS_USING_MCS28

PDSCH_TRANS_USING_MCS27





























Link Adaptation Measurements for DL PDSCH Usage per MCS over 1 hour 10,00,000

1,00,000

10,000

1,000

100

10

1


48




Link Adaptation for Signaling (PDCCH)

72 central subcarriers = 6 RBs (minimum LTE Bandwidth = 1.4 MHz)

1 CCE

Reservation of control Ch. Elements/RE

Bad air interface interference conditions limits available resources for PDCCH and impacts limits of PDSCH

= 36 RE

Example: 1 and 8 CCE's may be used for the transmission of one PDCCH.

2 CCE = 72 RE

eNodeB

CCE: Control Channel Element DCI: DL Control Information PHICH

PCFICH

PDCCH

RS

M8010C67 / Average PDCCH power per RE per DCI

Quality



Time

4 CCE

2 CCE

Channel

With low bandwidth no PDSCH allocation possible (macro cell condition)

8 CCE

PDCCH Format

# CCEs

# REs

# PDCCH bits

0

1

36

72

1

2

72

144

2

4

144

288

3

8

288

576

Number of CCEs for 1 PDCCH may change from 1 – 8 CCEs  channel conditions 50

RA41333EN160GLA0


RA41333EN160GLA0

49


Link Adaptation for Signaling (PDCCH)

These counters are updated with every subframe for which 1,2 or 3 OFDM symbols are allocated to the PDCCH

51

RA41333EN160GLA0


RA41333EN160GLA0

50


PDCCH Resources - The MaximumNumberOfOFDMSymbolsForPDCCH parameter defines how many OFDM symbols can be used. - Example shows dynamic case for MaximumNumberOfOFDMSymbolsForPDCCH=3 (yellow) - In RL30 selection between 1,2 or 3 symbols is dynamic based on actLdPdcch setting

maxNrSymPdcch LNCEL; 1..3; 1; 3 actLdPdcch LNCEL; false (0), true (1); false

52

M8011C59

Number of subframes with 1 OFDM symbol allocated to PDCCH

M8011C60

Number of subframes with 2 OFDM symbols allocated to PDCCH

M8011C61

Number of subframes with 3 OFDM symbols allocated to PDCCH

RA41333EN160GLA0


See also PDCCH link adaption

RA41333EN160GLA0

51


PDCCH Usage

PDDCH symbols usage 80,00,000 70,00,000

# of frames

60,00,000 50,00,000

40,00,000

Nbr of frames, 1 M8011C59

30,00,000


20,00,000


10,00,000 0

53

RA41333EN160GLA0


RA41333EN160GLA0

52


Link Adaptation for Signaling (PDCCH) -

Similar to data transmission, signaling (PDCCH) must be robust enough for UEs with low SINR (e.g. at cell edge)

-

Transmission with low Effective Coding Rate ECR •

Increased resource utilization

•

Lower number of scheduled UEs

•

UEs with good SINR should occupy less PDCCH resources and operate with lower number of CCEs (higher ECR)

-

Link Adaptation must deal with a trade-off between signaling robustness (coverage) and signaling capacity

-

The feature must be enabled with the parameter enableAmcPdcch

• Macro cell case • Uniform UE distribution

enableAmcPdcch LNCEL, true/false, true

8 CCE

54

RA41333EN160GLA0

4 CCE

2 CCE

1 CCE


RA41333EN160GLA0

53


Link Adaptation for Signaling (PDCCH) Updated when Aggregation Groups1,2.3.4 are used for PDCCH scheduling.

8 CCE

55

RA41333EN160GLA0

4 CCE

1 CCE

2 CCE


RA41333EN160GLA0

54


Link Adaptation for Signaling (PDCCH) • PDCCH utilization -

M8011C39/C40/C41/C42: AGG1 to AGG4 useage

-

M8011C43/C44/C45/C46: AGG1 to AGG4 blocked

Aggregation group PDCCH utilization/week 25,00,00,000

20,00,00,000

AGG4 AGG3

0

AGG1, used AGG2, used AGG4, used AGG8, used

AGG2

AGG4 AGG3

10,00,00,000 5,00,00,000

AGG2

AGG2

AGG3

AGG4

15,00,00,000

400

Aggregation group PDCCH blocked/week

350 300 250

AGG1

AGG1

AGG1

200 150 100 50

0

Measurement period

56

RA41333EN160GLA0

AGG1, blks

AGG2, blks

AGG4, blks

AGG8, blks


RA41333EN160GLA0

55


Link Adaptation Measurements for DL • Performance on RLC level • If layer 1 re-transmission fails, packets will be affected on higher protocol level -

M8001C137: Number of first RLC transmissions on DL

-

M8005C138: Number of RLC retransmissions on DL

M8001C137 + M8001C138 Total RLC packets

57

M8001C138 Retransmitted RLC packets

RA41333EN160GLA0

No counter Discarded RLC packets


RA41333EN160GLA0

56


Link Adaptation KPIs for DL LTE_5208a E-UTRAN RLC PDU Re-transmission Ratio Downlink Shows the retransmission ratio for RLC PDUs in downlink direction.

Logical formula


DL RLC PDU ReTrR = (number of retransmitted RLC PDUs) / (number all trans RLC PDUs)

Sum ([M8001C138]) Sum ([RLC_PDU_RE_TRANS]) / / sum sum ([RLC_PDU_FIRST_TRANS] + ([M8001C137] + [RLC_PDU_RE_TRANS]) * 100 [M8001C138])*100

58

RA41333EN160GLA0



RA41333EN160GLA0

57


E-UTRAN RLC PDU Re-transmission Ratio 0.30

0.25

0.20

0.15

0.10

0.05

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

0.00

LTE_5207a RLC PDU Re-transmission Ratio Uplink (%) LTE_5208a RLC PDU Re-transmission Ratio Downlink (%)

59

RA41333EN160GLA0


RA41333EN160GLA0

58



60

RA41333EN160GLA0


RA41333EN160GLA0

59


Inner Loop Link Adaptation (UL) Inner loop link adaptation adjust UE service close to BLER target

BLER target

BLER = f (SINR)

Actual BLER

Actual MCS will be performed by difference of actual BLER/target BLER

PwR

Actual BLER is estimated by ACK/NACK feedback from L1/L2 Orig i

MCS 20

5/5

MCS 16

nal Tran s

ACK / nR

NAC

etra

64QAM

nsm

mis sion

K

issi

ons

MCS 9 2/5

16QAM QPSK

1/5 Link adaptation

eNodeB UL Adaptive Modulation & Coding shall select its scheme table according to the radio conditions

61

RA41333EN160GLA0


RA41333EN160GLA0

60


Inner Loop Link Adaptation • Shall maintain for the UE a BLER close to the BLER target defined by the parameter • ulTargetBler

• User data and L3 signaling multiplexed together on PUSCH have a common BLER Target • The actual BLER is estimated on the basis of the ACK/NACK feedback obtained from L1/L2 • Inner loop LA will be performed every time the timer ulamcSwitchPer expires (i.e. if a certain • number of transport blocks has been sent) • Based on the difference actual BLER - target BLER the actual MCS will be upgraded or • downgraded

ulamcSwitchPer

ulTargetBler LNCEL, 10…50% with steps of 1%, 10%

62

RA41333EN160GLA0

LNCEL, 10…500 TBs with step of 10, 30 TBs


RA41333EN160GLA0

61


Upgrade / Downgrade of MCS • Each service starts with an initial coding scheme defined by the parameter iniMcsUl • Upgrade to faster MCS performed, if BLER better than ulTargetBler / ulamcUpdowngrF • Downgrade to slower MCS performed, if BLER worse than ulTargetBler * ulamcUpdownF • Upgrade and downgrade always performed by single MCS step

iniMcsUl

LNCEL, 0..20, 5 ulamcUpdowngrF UL AMC/BLER MCS performance

LNCEL, 1…3 with step of 0.05, 1.2

Low BLER 10% / 1.2 = 8.3% Rounded to 8%

3 2 Step ranges

1

High BLER 10% * 1.2 = 12% Rounded to 12% Low throughput

Upgrade MCS

Downgrade MCS Target BLER (10%)

63

RA41333EN160GLA0

Maintain MCS


RA41333EN160GLA0

62


MCSs for UL MCS_ index 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

ITBS

Mod order

0 1 2 3 4 5 6 7 8 9 10 10 11 12 13 14 15 16 17 18 19 19 20 21 22 23 24 25 26

2 2 2 2 2 2 2 2 2 2 2 4 4 4 4 4 4 4 4 4 4 6 6 6 6 6 6 6 6

reserved

Redundancy Version 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3

- UL AMC shall select the MCS to be employed from the tables according to the radio conditions - Early Releases: Range MCS 0..20. - LTE829 (RL30) allows for extending the range of MCSs used for 16QAM UEs to MCS 24  Approximately 25% higher UL peak rates compared to MCS 20.

- LTE44: Uplink 64 QAM (FL16) introduces 64 QAM modulation scheme in UL increasing maximum achievable UE uplink throughput in very good radio conditions and improving average cell capacity. Related Counters -also available with FL16- are: PUSCH_1ST_TRANS_MCS25 (M8001C274) .. 64QAM 16QAM QPSK 2 bits/symbol

b0 b1b2b3

b0 b1

00

Im

Im

64QAM b0 b1b2 b3 b4 b5 Im

16QAM

QPSK

01

6 bits/symbol

4 bits/symbol

1111

11

Re

10Re

Re

0000

64

RA41333EN160GLA0


RA41333EN160GLA0

63


Outer Loop Link Adaptation (UL) Inner loop link adaptation can be interrupted by outer loop link adaptation

M 8005 MCS upgrades / downgrades M8001 BLER (acks/nacks)

Inner loop link adaptation adjust UE service close to BLER target

Outer loop link adaptation based mainly on retransmission ACK/NACK & BLER counts

High SINR Low BLER

BLER = f (SINR) Actual BLER increase

BLER target

Actual BLER

Link adaptation

BLER target

high MCS coding

Low SINR High BLER

eNodeB OLLA based on actual BLER/MCS 65

RA41333EN160GLA0

lower MCS more convolution/retransmissions


RA41333EN160GLA0

64


Emergency Downgrade / Fast Upgrade of MCS (UL) -

-

Based on the ACK/NACK feedback of first transmission a correction ∆C is estimated Fast upgrade • With each ACK, ∆C is incremented by hardcoded step • MCS upgrades / downgrades • If ∆C exceeds hardcoded maximum, fast upgrade is done - M8005C140: Number of upgrades Emergency downgrade - M8005C141: Number of downgrades • With each NACK, ∆C is decremented by hardcoded step • If ∆C exceeds hardcoded minimum, emergency downgrade is done

OLLA Comp. C Max Cmax

FUG Event

Reduced AMC Period

0

Time

AMC Switching Period

EDG Event

Min Cmin 66

RA41333EN160GLA0

Reduced AMC Period


RA41333EN160GLA0

65


Link Adaptation Measurements (UL) •MCS usage -

M8001C16..C44: Histogram for number of PUSCH transmissions with MCS0… MCS28

-

M8001C74..C102: Histogram for number of not acknowledged PUSCH transmissions with MCS0..MCS28

-

M8001C177..C197, 485..488: Histogram for discarded PUSCH transmissions with MCS0..MCS28 due to maximum number of retransmissions

-

M8001C435..459: First transmissions on PUSCH using MCS0.. MCS24

-

M8001C460..484: First transmission NACKs on PUSCH using MCS0..MCS24

# of MCS counts

M8001C16..C44 Total transmission

M8001C74. . C102 Transmission with NACK

M8001C435 .. C459 First transmission

M8001C74. . C102 First transmission With NACK

M8001C177..C197 and 485 ..488 Discarded transmission

Expired maximum retransmissions cycles

Total counts for PUSCH allocation

67

Negative counts for PUSCH (non-actual allocation)

RA41333EN160GLA0

Negative counts for PUSCH non-allocation


RA41333EN160GLA0

66


Link Adaptation Measurements for UL • Performance on RLC level • If layer 1 re-transmission fails, packets will be affected on higher protocol level -

M8001C142: Total number of received RLC packets on UL

-

M8001C143: Number of received duplicate (= retransmitted) packets on UL

-

M8001C145: Number of discarded RLC packets on UL

# of MCS Counts M8001 Counter: M8001C142

Total RLC packets

68

RA41333EN160GLA0

Counter: M8001C143

Counter: M8001C145

Retransmitted RLC packets

Discarded RLC packets


RA41333EN160GLA0

67


Link Adaptation KPIs for UL LTE_5207b E-UTRAN RLC PDU Re-transmission Ratio Uplink Shows the retransmission ratio for RLC PDUs in uplink direction.

Logical formula UL RLC PDU ReTrR = (number of received duplicated RLC PDUs) / (number all received RLC PDUs)

69

RA41333EN160GLA0



Sum ([M8001C143]) / sum ([M8001C142]) * 100

Sum ([RLC_PDU_RE_TRANS]) / sum ([RLC_PDU_FIRST_TRANS] + [RLC_PDU_RE_TRANS]) * 100


RA41333EN160GLA0

68



70

RA41333EN160GLA0


RA41333EN160GLA0

69


Adaptive Transmission Bandwidth UL AMC performs adaptive transmission bandwidth (ATB)

• ATB is used due to limited UE power. • Less physical resource blocks, allowing less data transmission at cell edge, increases power per Hertz

• Shaded area = power • If BW is reduced power per hertz can be increased

power

UE allocated bandwidth 71

RA41333EN160GLA0


Besides selecting the most appropriate MCS according to radio conditions, the UL AMC shall also perform slow adaptive transmission bandwidth (ATB) ATB is necessary in case of lack of UE power to concentrate the remaining power on less physical resource blocks, allowing a regular data transmission in UL even at cell edge ATB informs the scheduler about the maximum number of resource blocks per TTI that can be assigned to a UE The decision about ATB is based on power headroom reports of the UE ATB is carried out, if a certain number of MCS upgrades or downgrades defined by the parameter ulatbEventPer has been done by the inner or outer loop link adaptation

RA41333EN160GLA0

70


Power Headroom Reporting Head room counter M8005 C54 … C85 The decision about ATB is based on power headroom reports of the UE > 39 dB power headroom -23..-21 dB power headroom

PwR (dBm)

MME M 8005 PwR headroom UE reports to NW periodically

Service UL PwR for service must increase due to high path loss!

Max Path loss

PwR loss

Service

cell cell cell

Last counter of power headroom

72

Event based trigger in case of: DL Path loss change is bigger than a defined threshold, reported by UE to eNodeB

RA41333EN160GLA0

1.st counter of power headroom

M 8005 PwR headroom measurements


Purpose ATB is to provide RB for UL. It is completely different from DL as there is no Power C in DL but in UL. In RL20:So the less power the UE has, the less RB will be provided in order maintain the SINR at the eNB. Now in RL30 with EULA it has changed due to Shannons theorem, the bandwidth is maintained for longer and the MCS is downgraded first and then the bandwidth is reduced. In Rl60 the FULA algorithm further enhances the EULA algorithm.

RA41333EN160GLA0

71


Power Headroom Measurements Measurements for PUSCH

MME

- M8005C87, C88, C89: Minimum/maximum/mean power headroom - M8005C54..C85: power headroom histogram

PH (i)  PCMAX  10 log10 (M PUSCH (i))  PO_PUSCH ( j )    PL  TF (i)  f (i) dB M 8005 PwR headroom

cell cell

Power headroom values

cell

73

Max

RA41333EN160GLA0

>39dB

…………..

37 to 39 dB

35 to 37dB

-17 to -15dB

-19 to -17 dB

-21 to -19dB

Min

-23 to -21 dB

Mean


PPUSCH (i) :PUSCH Power in subframe i PCMAX: max. allowed UE power (23 dBm for class 3) MPUSCH: number of scheduled RBs (The UE Tx. Power increases proportionally to # of PRBs) PO_PUSCH(j) = PO_NOMINAL_PUSCH(j) + PO_UE_PUSCH(j) PL: pathloss [dB] = referenceSignalPower – higher layer filtered RSRP DTF (i) = 10 log 10 (2MPR Ks – 1) for Ks = 1.25 else 0, MPR = TBS/NRE, NRE : number of RE Ks defined by deltaMCS-Enabled, UE specific f(i): TPC (Closed Loop adjustment) Semi-persistant: j=0 / dynamic scheduling: j=1 PO_NOMINAL_PUSCH(0,1): cell specific (SysInfo) PO_UE_PUSCH(0,1): UE specific (RRC) a (0,1) = 0.0, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0 (partial PL compensation by open loop) Random access grant: j=2 PO_NOMINAL_PUSCH(2): PO_PRE + DPreamble_Msg3 PO_UE_PUSCH(2) = 0 a (2) = 1.0 (i.e. full PL compensation)

RA41333EN160GLA0

72



74

RA41333EN160GLA0


RA41333EN160GLA0

73


DL Scheduling • LTE downlink scheduler is channel aware and assigns resources in a proportional fair manner • Resource allocation is done in time domain as well as frequency domain • Specific priority shall be granted to Signaling Data and to HARQ retransmission • QoS support is based just on minimum and maximum bit rate control per UE. Additionally UE capabilities are supported • The Number of UEs simultaneously scheduled per TTI can be limited by the parameter maxNumUeDl • The maximum number of UE’s that can be scheduled per TTI depends on the bandwidth employed

• Scope of the packet scheduler is cell level

maxNumUeDl LNCEL,

1..7 for 1.4 MHz BW, 7 1..7 for 3 MHz BW, 7

FDD

maxNumUeDl LNCEL,

TDD

1..10 for 10 MHz BW, 10 1..12 for 20 MHZ BW, 12

1..7 for 5 MHz BW, 7 1..14 for 10 MHz BW, 14 1..17 for 15 MHz BW, 17 1..20 for 20 MHZ BW, 20

75

RA41333EN160GLA0


RA41333EN160GLA0

74


DL Scheduling - Finally allocate UEs (bearers) onto PRB, considering only the PRBs available after the previous steps • Pre-Scheduling: All UEs with data available for transmission based on the buffer fill levels • Time Domain Scheduling: Parameter maxNumUeDl decides how many UEs are allocated in the TTI being scheduled • Frequency Domain Scheduling for candidate set 2 UEs: Resource allocation in frequency domain including number and location of allocated PRBs

Start Pre-Scheduling: Select UEs eligible for scheduling -> Determination of Candidate Set 1 Time domain scheduling of UEs according to simple criteria -> Determination of Candidate Set 2 Frequency domain scheduling of UEs/bearers -> PRB/RBG allocation to UEs/bearers End

76

RA41333EN160GLA0


RA41333EN160GLA0

75


Measurements Related to DL Scheduling • Physical resource block utilization (= air interface load) -

M8011C35/C36/C37: Minimum/ maximum/mean % of occupied PRBs per TTI

-

Mean value stored at 10x greater i.e. no decimal point, example: 5.4 stored as 54

-

M8011C25..C34: Histogram for % of occupied PRBs per TTI The histograms work as follows 1.st counter: Number of measurements with utilization < 10% 2.nd counter: Number of measurements with utilization between 10..20% Continue with steps of 10%

% occupied PRB / TTI

Last counter: Number of measurements with utilization > 90%

77

RA41333EN160GLA0

90% to 100% utilization








Min


Mean

<10% utilization

Max


RA41333EN160GLA0

76


KPIs Related to DL Scheduling LTE_5276b E-UTRAN average PRB usage per TTI DL Shows the average value of the Physical Resource Block utilization per TTI in DL direction The utilization is defined by the ratio of used to available PRBs per TTI

Logical formula



AVG DL PRBs = (average DL PRBs per TTI)

Avg ([M8011C37])/10

Avg ([DL_PRB_UTIL_TTI_MEAN])/10

-

78


RA41333EN160GLA0


RA41333EN160GLA0

77


CMAS ETWS PM Counters Several counters are introduced to support LTE494 Commercial Mobile Alert System (RL40) and LTE843 ETWS broadcast (RL40/RL35TD) Full name

Measur Used Used ement by by level CMAS ETWS MME per eNB

Yes

Yes

M8000C40

Number of sent S1-AP S1AP_WRITE_REP_ WRITE-REPLACE WARN_RESP WARNING RESPOPNSE

per eNB

Yes

Yes

M8008C16

RRC_PAGING_ETW Number of RRC-Pagings due per cell S_CMAS to CMAS/ETWS

Yes

Yes

M8000C41

S1AP_KILL_REQ

Number of received S1-AP KILL REQUEST

per eNB

Yes

No

M8000C42

S1AP_KILL_RESP

Number of sent S1-AP KILL RESPONSE

per eNB

Yes

No

M8001C231

NUM_WARN_ETWS Number of SIB10-broadcasts per cell _PRIM

No

M8001C232

NUM_WARN_ETWS Number of SIB11-broadcasts per cell _SEC

No

Yes

M8001C233 NUM_WARN_CMAS Number of SIB12-broadcasts per cell

Yes

No

WRITEREPLACE WARNING REQUEST

Number of received S1-AP S1AP_WRITE_REP_ WRITE-REPLACE WARN_REQ WARNING REQUESTS

KILL REQUEST (CMAS only)

M8000C39

WRITEREPLACE WARNING RESPONSE

NetAct name

KILL RESPONSE (CMAS only)

CounterID

KILL messages are used to abort the broadcast of a warning message

Yes

eNodeBn

UE

UE UE

79

RA41333EN160GLA0


RA41333EN160GLA0

78


LTE1800: Downlink interference shaping Introduction: Fractional load and CQIs frequency

Traffic pattern in the cell – the load is fractional, but now allocations are placed only within preferred area

This profile of traffic creates spatially localized interference to the neighboring cell

Preferred allocation area

Squeezing the traffic to a section of the bandwidth we create distinct “busy” and “clean” areas

I o u r

Area excluded from allocations

time

80

RA41333EN160GLA0

Now the subband CQIs will be able to well depict the interference pattern generated by neighboring cell


RA41333EN160GLA0

79


LTE1800: Downlink interference shaping

actDlIntShaping Activate downlink interference shaping LNBTS; False, True

Fractional loaded Cell to Highly loaded Cell

The simplest scenario where this feature is expected to provide a gain is a fractionally loaded cell with a neighboring highly/fully loaded cell.

Heavy load

CQI

Fractional load

f

Cell1 Cell2

PRB utilization

f

81

RA41333EN160GLA0

Utilization of PRB resources in cell 1 becomes visible in the frequency selective channel quality reporting of a cell edge UE in cell 2.


actDlIntShaping activates the feature within the eNB actdLIsh activates feature on Cell level

RA41333EN160GLA0

80


LTE1800: Downlink interference shaping Performance Management

Number

Abbrev name

Full name

M8011C87

DL_INTERFER_SHAP_USE

DL interference shaping usage

M8011C88

DL_INTERFER_SHAP_AMOUNT

Amount of preferred resources per TTI

M8011C89

DL_INTERFER_SHAP_CHANGE

Number of changes of preferred resources

82

RA41333EN160GLA0


RA41333EN160GLA0

81


LTE1113 /LTE1496 eICIC What is eICIC? Enhanced Inter Cell Interference Co-ordination Rel10

Co-ordination On X2

Why eICIC? • Small Cell in Macro cell coverage suffers low SINR • solution – prevent Macro from Txing on specific subframes • reduction in Macro capacity compensated by increase in Small cell capacity

83

RA41333EN160GLA0


Placing small-cells within the coverage area of macro eNBs introduces several challenges in terms of interference management. The perceived Signal to Interference and Noise Ratio (SINR) by pico-UEs in the extended area is poor, due to the stronger macro interference and lower signal strength from the serving picoeNB.

RA41333EN160GLA0

82


LTE1113 /LTE1496 eICIC Muted subframes known as ABS (Almost Blank)

During ABS-frames small cells can serve cell edge mobiles

SMALL cell transmission subframes During all subframes good radio condition mobiles served

Ma

cro

-int e

rfe re

n ce na

l

MACRO cell transmission subframes

si g

Range extension

pico-eNB

M8011C112 M8011C113 M8011C114 M8011C115 M8011C116 M8011C117

macro-eNB UE

X2-link with eICIC coordination

84

RA41333EN160GLA0

 with eICIC larger Range Extension values can be applied + better conditions for small cell edge camped mobiles E_ICIC_MUTING_PATTERN_1 E_ICIC_MUTING_PATTERN_2 E_ICIC_MUTING_PATTERN_3 E_ICIC_MUTING_PATTERN_4 E_ICIC_MUTING_PATTERN_5 E_ICIC_MUTING_PATTERN_6

M8011C118 –C139 eICIC UL/Dl PRB utilization level © Nokia Solutions and Networks 2016

The eICIC functionality will establish eICIC partnerships between a macro eNB and its sub-tending pico eNBs •Operator configures one macro cell partner on each pico. •A pico cell can only have one macro cell partner. •Following X2 setup, if a pico is configured with a macro cell partner the pico will initiate the eICIC partnership establishment by sending an X2: LOAD INFORMATION message to the macro.

RA41333EN160GLA0

83


LTE1113 /LTE1496 eICIC ABS muting patterns FN subframe

M8011C112-117 eICIC accumlated usage muting pattern 1-6 M80011C164 eICIC accumlated usage muting pattern 0

PSS/SSS PBCH SIB1 PCH HARQ 0 HARQ 1 HARQ 2 HARQ 3 HARQ 4 MP

85

RA41333EN160GLA0

0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9

1

MP ratio 0,15

0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0

2

0,25

0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0

3

0,375

0 1 1 0 0 1 0 0 0 1 1 0 0 1 0 0 0 1 1 0 0 1 0 0 0 1 1 0 0 1 0 0 0 1 1 0 0 1 0 0

4

0,5

0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0

5

0,625

0 1 1 1 0 1 1 0 0 1 1 1 0 1 1 0 0 1 1 1 0 1 1 0 0 1 1 1 0 1 1 0 0 1 1 1 0 1 1 0

6

0,75

0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1 1


MP= Muting pattern

RA41333EN160GLA0

84


LTE1113 /LTE1496 eICIC Static - eICIC Cell Range Expansion (CRE) • eICIC uses Cell Individual Offsets to realize small cell range expansion

• cell range expansion helps to move traffic load to small cells

M8011C118-128 number of DL ABS SF with UP traffic 0-100% M8011C129-139 number of UL ABS SF with UP traffic 0-100%

Micro

Macro

RA41333EN160GLA0

86

RSRP of serving Macro

ABS SF UP traffic of CRE Ues90 -100%

ABS SF UP traffic of CRE Ues 80-90%



ABS SF UP traffic of CRE Ues 0%

ABS SF UP traffic of CRE Ues 010%

A3 trigger with CRE RSR P

A3 trigger w/o CRE

RSRP of neighbor Small Cell

CR E

A3 offset

t


due to low transmit power small cells have typically low coverage, resulting in low number of captured macro UEs

• small cell range can be extended via cell individual offsets: ‑ ‑

cellIndOffServ: cell-individual offset of serving cell (Ocp) cellIndOffNeigh: cell-individual offset of neighbor cells (Ocn)

• with proper settings of cell individual offsets it is possible to: ‑ optimize intra-LTE handovers •

i.e. A3 or A5 HO to small cell may be triggered faster

‑ realize of particular mobility management strategy •

load balance between Macro and Small Cell

RA41333EN160GLA0

85


UL Scheduling • UL packet scheduling is responsible for allocating PUSCH resources (PRBs) to UEs in time and frequency domain

• In UL (unlike DL) only contiguous allocation of PRBs to a UE possible • One uplink scheduler per cell, separate schedulers for downlink and uplink • Channel unaware scheduler, interference aware scheduler and channel aware scheduler are available to be used in the frequency domain • There is no MIMO support in the UL-Scheduler

• The Number of UEs simultaneously scheduled per TTI is limited by the parameter maxNumUeUl (analogous to the DL)

maxNumUeUl LNCEL,

1..7 for 1.4 MHz BW, 7

FDD

maxNumUeUl LNCEL,

1..7 for 3 MHz BW, 7

1..10 for 10 MHz BW, 10

TDD

1..12 for 20 MHZ BW, 12

1..7 for 5 MHz BW, 7 1..10 for 10 MHz BW, 14 1..15 for 15 MHz BW, 17 1..20 for 20 MHZ BW, 20

87

RA41333EN160GLA0


RA41333EN160GLA0

86


LTE1117: MBMS

1. evolved Multimedia Broadcast Multicast Services (eMBMS) is a technique to broadcast transmissions to a wide number of users, located within a confined geographical area 2. Complete TTIs are reserved for MBMS transmissions, exhausting all PRBs within this TTI

3. “Reserved TTIs” ranges from 0 – 60%, in 2.5% increments 4. Gain increases significantly with increasing number of users Compared to redundant, unicast transmissions LNBTS: actMBMS Activate support for MBMS 0 (false), 1 (true) Default: 0

5. Typical use case: Sports events

88

RA41333EN160GLA0


RA41333EN160GLA0

87


LTE1117: MBMS MBMS architecture Video Content Provider

BM-SC IMS

Internet

Schedule Sessions

BM-SC SGi

Identify Service Area

MBMS-GW

Request QoS for Session

IP Multicast M1

S-GW

Perform Synchronization SGi-mb

SGmb

WAP, http, MMS

P-GW

SGmb

Sm

Service Announcement Function

Initiate/Terminate MBMS Bearers

SGi-mb

HSS

Add Session Id

SGi

MME MBMS-GW S-GW Relay Signaling

M3

IP Multicast Mgmt

To UE

IP Multicast Transmission

MME MCE

Relay Signaling

MCE

MCE

eNB

M2 (internal)

MCE Admission Control

eNB

eNB X2

X2

Scheduling

eNB BM-SC MBMS GW MCE

89

Broadcast Multicast Service Center MBMS Gateway Multi-cell/Multicast Coordination Entity RA41333EN160GLA0

Multicast Bearer Mngmt

Multicast Bearer Xmission


The Flexi Multiradio BTS supports the MBMS by adding the multicell/multicast coordinating entity (MCE) to the existing functionality. The MCE is a software entity integrated in the eNodeB, and it follows a decentralized MCE architecture. The MCE takes care of MBMS resource management and control information scheduling across cells.

RA41333EN160GLA0

88


LTE1117: MBMS Performance Management New counter Group M8030 Number

Abbrev name

Full name

M8030C1

MBMS_SESSION_ACT_AVG

Average number of activated MBMS sessions

M8030C2

MBMS_USER_DATA_EUU

MBMS user data volume (eUu interface)

M8030C3

MBMS_USER_DATA_M1

MBMS user data volume (M1 interface)

M8030C4

MBMS_USER_DATA_M1_LOST

Lost MBMS user data volume (M1 interface)

M8030C5

MBMS_USER_DATA_DROP_1

Dropped MBMS user data volume 1 (M1 interface)

M8030C6

MBMS_USER_DATA_DROP_2

Dropped MBMS user data volume 2

90

RA41333EN160GLA0


RA41333EN160GLA0

89


UL Scheduling - In UL only contiguous allocation of PRBs to a UE possible

DOWNLINK

- Channel unaware frequency scheduling Random allocation of PRBs to Ues.

- Channel aware frequency scheduling uses SRS.

UE1 UE2 SINR

- Interference aware scheduler allocates cell edge UE’s to PRB regions with low noise and interference.

Frequency UPLINK UE4 UE1 UE3 UE5 UE2

UE5 UE3 UE2

UE4 UE1

UE4

UE4 UE5 UE3

UE1

UE1 UE5

UE2

UE3

SINR

UE2 UE5 UE1 UE4 UE2

UE2 UE3 UE2 UE5

PRB

UE3 UE1

UE4 UE1

TTI

91

RA41333EN160GLA0

Time © Nokia Solutions and Networks 2016

RA41333EN160GLA0

90


UL Scheduling -

Evaluate number of PRBs that will be assigned to UEs

-

Evaluate available number of PRBs per user (resources are assigned via groups of consecutive PRBs

•Time domain -

Like on the DL the maximum number of UEs which can be scheduled per TTI time frame is restricted in dependence on the bandwidth

Frequency Domain -

Use a random function to assure equal distribution of PRBs over the available frequency range (random frequency hopping)

Example of allocation in frequency domain Full Allocation All available PRBs are assigned to the scheduled UEs per TTI

Fractional Allocation Not all PRBs are assigned Hopping function handles unassigned PRBs as if they were allocated to keep the equal distribution per TTI

92

RA41333EN160GLA0


RA41333EN160GLA0

91


Measurements Related to UL Scheduling • Physical resource block utilization (= air interface load) -

M8011C22/C23/C24: Minimum/maximum/mean % of occupied PRBs per TTI

-


-

M8011C12..C21: Histogram for % of occupied PRBs per TTI

-

The histograms work as follows 1.st counter: Number of measurements with utilization < 10% 2.nd counter: Number of measurements with utilization between 10..20% Continue with steps of 10%

% occupied PRB / TTI

Last counter: Number of measurements with utilization > 90%

93

RA41333EN160GLA0









Min


Mean

<10% utilization

Max


RA41333EN160GLA0

92


KPIs Related to UL Scheduling LTE_5273b E-UTRAN average PRB usage per TTI UL

-FDD

Shows the average value of the Physical Resource Block utilization per TTI in UL direction The utilization is defined by the ratio of used to available PRBs per TTI

Logical formula



AVG UL PRBs = (average UL PRBs per TTI)

Avg ([M8011C24])/10

Avg ([UL_PRB_UTIL_TTI_MEAN])/10

-

94


RA41333EN160GLA0


RA41333EN160GLA0

93


E-UTRAN average PRB usage per TTI 2.0 1.8 1.6 1.4 1.2 1.0

0.8 0.6 0.4 0.2

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

0.0

LTE_5273a Average PRB usage per TTI Uplink (%) LTE_5276a Average PRB usage per TTI Downlink (%)

95

RA41333EN160GLA0


RA41333EN160GLA0

94


LTE1059 Uplink multi-cluster scheduling (FL16/TL16) • In case when LTE944 PUSCH masking or LTE825 Outer region scheduling is used in the network, PUSCH spectrum is divided into several PUSCH areas • Divided PUSCH causes uplink peak UE throughput decrease when only one UE is scheduled in single TTI, because UE can be scheduled only in one PUSCH area at the time • LTE1059 Multi-cluster scheduling introduces possibility to schedule UE in two PUSCH areas at the same time when there is only one UE scheduled in single TTI Before: Single UE can be scheduled in only one PUSCH area in single TTI PUCCH

PUSCH area 1

PUSCH area 2

PUCCH

After: Single UE can be scheduled in two PUSCH areas in single TTI. Peak throughput is increased

Counter name

Description

MC_SUB_FRAME

This counter provides number of subframes with multi-cluster scheduling

(M8011C170)

Trigger event: This counter is incremented per sub-frame, if there is multi-cluster scheduling

#LTE Cell Resource

Use case: This counter can be used to derive usage of multi-cluster scheduling in terms of TTIs

96

RA41333EN160GLA0


RA41333EN160GLA0

95


LTE2462 Coordinate Scheduling for Beamforming Interference Avoidance (FL16/TL16) • CSBIA - Coordinate Scheduling for Beamforming Interference Avoidance • This feature is to avoid beamforming interference from intra-site neighbor cells for users in sector boundary area, by coordinating the resource allocation among those cells.

• The feature can bring significant gain to the throughput of Cell Edge UE. - Operator can adjust the throughput gain by tuning the CSBIA Region’s size - The feature can also bring extra throughput gain by avoiding the beamforming interference.

Counter name

Description

UE_CSBIA_DL_SUB FRAME_AVG (M8011C188)

This measurement provides the arithmetical average of samples taken from the number of CSBIA interfered UEs in DL Subframes. The reported value is 100 times higher than the actual value (for example 1.00 is stored as 100). Trigger event: This counter is updated by the arithmetical average of samples, each representing the number of CSBIA interfered UEs on the downlink at that moment. Granularity: 1s KPI: LTE_5796a

#LTE Cell Resource

97

RA41333EN160GLA0

Cell A

Cell B


RA41333EN160GLA0

96



98

RA41333EN160GLA0


RA41333EN160GLA0

97


2 Tx Diversity Benefit: Diversity gain, enhanced cell coverage • Each Tx antenna transmits the same stream of data  receiver gets copies of the same signal which increases the SINR

• Synchronization signals are transmitted only via the 1 st antenna • eNodeB sends different cell specific Reference Signals per antenna • The feature has to be enabled on cell basis by the parameter DlMimoMode • Processing is completed in 2 phases • Layer Mapping: distributing a stream of data into two streams • Pre-coding: generation of signals for each antenna port

FDD TDD DlMimoMode LNCEL, 0..60, 10 SingleTX (0), TXDiv (10), 4way TXDiv (11), Dynamic Open Loop MIMO (30), Closed Loop Mimo (40), Closed Loop MIMO (4x2) (41), Closed Loop MIMO (8x2) (42), Closed Loop MIMO (4x4) (43), Single Stream Beamforming (50), Dual Stream Beamforming (60)

DlMimoMode LNCEL, 0..41, 11 SingleTX (0), TXDiv (10), 4-way TXDiv (11), Dynamic Open Loop MIMO (30), Closed Loop Mimo (40), Closed Loop MIMO (4x2) (41)

…s2, s1

s1

s2* Antenna Port 0

Alamouti encoder

s2  s1* Antenna Port 1

99

RA41333EN160GLA0


RA41333EN160GLA0

98


TX diversity for 4 antennas

RL50

For 4 Tx ant, TX diversity uses combination of SFBC and FSTD

To balance for channel estimation accuracy • {s1, s2} are transmitted by antenna ports 0 and 2 • {s3, s4} are transmitted by antenna ports 1 and 3

s1 s2* 0 0 Antenna Port 0

0 0 s3 s4* Antenna Port 1

… s 4 , s3 , s2 , s1

Alamouti encoder

s2  s1* 0 0 Antenna Port 2

Weaker channel estimation for antenna ports 2 and 3 (only 2 symbols for RS per PRB per slot)

0 0 s4  s3* Antenna Port 3

100

RA41333EN160GLA0


• DL adaptive closed loop MIMO (4x2) supports Transmit Diversity for 4 antenna ports in Transmission Mode 4 (TM4) and in Transmission Mode 2 (TM2). • 3GPP has specified open loop Transmit Diversity using one codeword: Precoding Feedback and Rank Information is not required! • Transmit Diversity using 4 antenna ports is used whenever there is no valid, complete and consistent Channel State Information available as detected by eNodeB. • During Initialization when RRC setup is performed • No update of valid CSI reports for single layer (RI=1) and dual layer (RI=2) transmissions since a characteristic update time. • UE does not send valid reports (e.g. Category 1 UEs). Transmit Diversity for 4 antenna ports is implemented as a combination of SFBC (Space Frequency Block Coding) with FSTD (Frequency Switched Time Diversity).

RA41333EN160GLA0

99


2x2 MIMO (1 Code Word via 1 Antenna) Benefit: Doubles peak rate compared to 1 Tx antenna

Two code words (S1+S2) are transmitted in parallel to 1 UE  double peak rate

• Signal generation is similar to transmit diversity, i.e. layer mapping and pre-coding

• Can be open loop or closed loop depending whether the UE provides feedback

• Static spatial multiplexing with 2 code words

S2

• Supported physical channel PDSCH

S1 Layer Mapping

Pre-coding

Code word 1 Modulation

× L1

Scale

Modulation

× L2

101

RA41333EN160GLA0

Map onto Resource Elements

OFDMA



Map onto Resource Elements

OFDMA

W1

× Code word 2



×

W2


RA41333EN160GLA0

100


2x2 MIMO (1 Code Word via 2 Antennas) Benefit: High peak rates and good cell edge performance

1 code word A is transmitted via 2 antennas to 1 UE to improve the link budget

• Dynamic selection between transmit diversity and open loop spatial multiplexing with 2 code words based on CQI

• If disabled case either static spatial multiplexing or static Tx diversity can be selected for the whole cell only (i.e. for all UEs)

• Supported physical channel: PDSCH

A A B

2 code words (A+B) are transmitted in parallel to 1 UE which doubles the peak rate

LiBu: Link Budget

102

RA41333EN160GLA0


RA41333EN160GLA0

101


Dynamic Open Loop MIMO mode

CQI

SM

SM

mimoOlCqiThU mimoOlCqiThD Time

Pre RL50

Filtered: Filtere cqi, d ri

RI

Inactivity: Inactivit Aging y: aging applied

mimoOlRiThU mimoOlRiThD Time

From RL45 / RL50 Uses fast adaptive MIMO switch, uses the RI report directly

103

RA41333EN160GLA0


CQI Filter: - For Diversity (Report with Rank = 1) newCQI = WIDEBAND_CQI_Stream1 + DELTA_CQI*rdMimoOllaUsedXX. -For Spatial Multiplexing (Report with Rank = 2) newCQI = (WIDEBAND_CQI_Stream1 + WIDEBAND_CQI_Stream2) / 2 + mimoCqiCompSmDivXX + DELTA_CQI*rdMimoOllaUsedXX. During times on inactivity, also the CQI filter shall be slightly modified to provide a slow and seamless movement towards Diversity Mode independent from mimoXXCqiAvg settings. For every TTI mimoCQI(t) is decreased by: mimoCQI(t) = rdMimoCqiAgeing *mimoCQI(t-1) Averaging: mimoCQI(t) = (1-mimoXXCqiAvg)*mimoCQI(t-1) + mimoXXCqiAvg*newCQI.

RA41333EN160GLA0

102


CL Spatial Multiplexing Dual stream (4x2)

RL50

• 2 codewords only TM4 • Precoding w/o CDD • Code book based 16 index values as per 3GPP 36.211 R9, precoding matrix W: • UE feedback: precoding matrix indicator (PMI)

Codeword 0

Layer Mapper

Layer 0 Precoding

Layer 1 Layer 2

1 stream shown but 2 data streams are supported

Layer 3

Feedback: CQI RI PMI

No new counters or KPIs

104

RA41333EN160GLA0


2 codewords are the 3GPP max – Ack/Nck and CQI are per codeword – 2 CW gives an optimum overhead. Even with high order layers (say 8x8) still only 2 CW but we are sending the codewords much faster!

RA41333EN160GLA0

103


LTE569 4x4 MIMO

TL15a

MIMO Principle: 4x4 h11 h12

In case of 4x4 MIMO there are 4 TX antennas at the eNB, and 4 RX antennas in the UE

h13 h14

y1  h11x1  h12 x2  h13 x3  h14 x4 y2  h21x1  h22 x2  h23 x3  h24 x4 y3  h31x1  h32 x2  h33 x3  h34 x4 y4  h41x1  h42 x2  h43 x3  h44 x4

x1

TX antenna port 0

h21

Rx1

y1

h22

xx2 2

TX antenna port 1

h23 h24

x1 Rx2

h31

xx33

Precoder

TX antenna port 3

y3

h32 Rx3

h33

Layers

xx44 …to antenna ports

y2

TX antenna port 4

MIMO receiver

4 equations, 4 unknowns: more complexity needed in the UE

x2 x3 x4

y4

h34 h41

Rx4

h42 h43 h44

105

RA41333EN160GLA0


RA41333EN160GLA0

104


LTE569 4x4 MIMO

TL15a

M8001C234 ACTIVE_UE_CAT_8_AVG M8001C493 ACTIVE_UE_CAT_5_AVG M8010C57 MIMO_CL_1CW M8010C58 MIMO_CL_2CW M8010C71 MIMO_CL_2CW_3LAYER

Category 5 or 8 UEs are required to fully benefit from this feature. Legacy UEs will be scheduled using 4x2 MIMO schemes.

M8010C72 MIMO_CL_2CW_4LAYER M8010C73 MIMO_CL_1CW_2LAYER

106

RA41333EN160GLA0


RA41333EN160GLA0

105


Measurements Related to MIMO • Usage of transmission modes

MME

M8010C55: MIMO Open Loop Diversity Mode M8010C56: MIMO Open Loop Spatial Multiplexing

M 8010 MIMO transmission on air interface

M8010C57: MIMO Closed Loop Single Codeword M8010C58: MIMO Closed Loop Double Codeword M8010C59: Open Loop MIMO mode switches

RAN QoS

M8010C60: Closed Loop MIMO mode switches

• MIMO Rank Indication reported by the UE* cell

M8010C112: UE reported RI 1

cell


cell

SAE GW


MME *Note, these counters are available from FL16/TL16 onwards

S1 signaling eNB

RAN QoS M 8010 CQI eNB DL air interface

M 8010 Air interface QoS 107

RA41333EN160GLA0


RA41333EN160GLA0

106


MIMO layer usage

Open loop MIMO 70.00 60.00

50.00

100.00

8,000

800

90.00

7,000

700

80.00

600

70.00

40.00

500

30.00

400

OL div OL spatial mux

300 200

10.00

100 0 11:00:00 13:00:00 15:00:00 17:00:00 19:00:00 21:00:00 23:00:00 01:00:00 03:00:00 05:00:00 07:00:00 09:00:00

0.00

RA41333EN160GLA0

MIMO mode switches

6,000

60.00

5,000

50.00

4,000

CL, Single Codeword

40.00

3,000

CL, Double Codeword

30.00

2,000

20.00

MIMO mode switches

1,000

10.00 0.00

0 12:00:00 13:00:00 14:00:00 15:00:00 16:00:00 17:00:00 18:00:00 19:00:00 20:00:00 21:00:00 22:00:00 23:00:00 00:00:00

20.00

108

Closed loop MIMO

900


RA41333EN160GLA0

107


TDD Beam Forming 1. BF: sharper beam with higher antenna gain • => more Tx power towards wanted UE • => coverage gain 

RL25 2. BF algorithms in RL15 don’t explicitly consider interference • lower I if beam doesn’t hit victim UE  • higher I if beam hits the victim UE  • overall no impact on interference 

interfered UE

interfered UE (served by another cell)

4. only 1 beam per frequency and time resource (no spatial multiplexing) => only moderate capacity gain by BF 

3. beam follows served UE  => very unstable interference  (flashlight effect) 109

RA41333EN160GLA0


RA41333EN160GLA0

108


TDD Beam Forming Dual Stream beamforming Dynamic Transmission Mode Switching

RL35

- RL35 introduce enhanced support for switching with TM3 and TM8 - a cell is configured to one of these modes: TM1, TM2, TM3, TM4, TM7, TM8 • TM1 to TM4 on sector beam (no beamforming) M8010C63 - switching inside TM7: TxDiv, SS-BF M8010C64 - switching inside TM8: TxDiv, SS-BF, DS-BF - RL35 switching from TM7 or TM8 to TM3 allowed

UE category =1

cell configured as TM7 cell configured as TM8

TM2

TM8 dual stream Beamforming mode TM8 single stream Beamforming mode

UE category > 1 UEs supporting UEs supporting TM7 (but not TM8) TM8 TM7 TM7 TM8

- retransmissions • always same rank and tx scheme as respective 1 st tx (SS-BF, DS-BF, TxDiv) • TM must not have changed between 1 st tx and re-tx 110

RA41333EN160GLA0


RA41333EN160GLA0

109


TDD Beam Forming RL25 Dual Stream beamforming LTE541 TxDiv fall back inside TM8 (based on CQI and rank)

rank switching M8010C65 TM8 transmit diversity mode

fall back

-

decision based on current mimoCQI and mimoRANK red region: use DS-BF (independent of current mode) yellow region: use SS-BF (independent of current mode) green region: use TxDiv (independent of current mode) blue region: if current mode = DS-BF: switch to SS-BF else: keep current mode (SS-BF or TxDiv) - orange region: keep current mode

111

RA41333EN160GLA0


RA41333EN160GLA0

110


TDD Beam Forming Dual User Single Layer MU-MIMO LTE1169

RL35

• TM8 (dual layer beamforming)-based MU-MIMO for two users and each user with only one layer. • Two users are paired and then share the same PRBs, - Each user one codeword • Uses TM8 antenna ports 7 and 8 • Each user mapped to different Antenna port • TM8 UEs can move between - SU single layer beamforming mode (TM8, single CW, SU-MIMO mode) - SU dual layer beamforming mode (TM8, two CW, SU-MIMO mode) - DU single layer beamforming mode (TM8, single CW, MU-MIMO mode)

M8010C66 TM8 dual-user single-stream Beamforming mode

112

RA41333EN160GLA0


RA41333EN160GLA0

111


TDD Beam Forming RL45 8x2 Single User MIMO with TM9 LTE1543 •8x2 MIMO over 8 TX pipes and 2RX UE antennas with closed-loop CSI feedback

Preferred beam information is signaled to eNB in an improved PMI report

Rel. 10 CSI-RS are sent in DL to help with channel estimation

CQI PMI Rank

•Dual stream transmission with TM9 for Rel. 10 TM-9 enabled UEs

CSI-RS

DM-RS

CRS

DL feedback

0 1 2 3 4 5 6 0 1 2 3 4 5 6

PDSCH PDCCH CRS DM-RS CSI-RS

MCS Rank UL feedback M8010C68 M8010C69

Improved PDSCH beam enhances DL SINR allowing for higher DL throughput

113

• • • •

RA41333EN160GLA0

M8010C70

TM7 Beamforming mode TM7 TxDiv mode TM9 dual-user single-stream mode


PDSCH channel estimation based on demodulation reference signals (DM-RS) – ports 7..14 (up to 8 layers) CQI report based on Channel State Information Reference Signals (CSI-RS) – ports 15..22 Closed loop precoding based on CSI-RS, reported by PMI CSI-RS is configured to each Rel10 UE via RRC reconfiguration

RA41333EN160GLA0

112


LTE1787 TMP with 8Tx MU MIMO

TL15a

• LTE 1787 allows to transmit signal to 2 users (downlink) at the same time and in the same frequency resources by using beamforming and spatial multiplexing. • It is a replacement for LTE1169 ”TM8 based Dual User Single Layer MU-MIMO”. • It supports transmission mode 9 (TM9) single layer MU-MIMO by reusing and adapting algorithms of LTE1169. UE1 gets 9 RBGs multiplexed with UE2

UE1

eNB RBGs

UE1

UE3

UE2

114

8 TX (4 X-POL)

RA41333EN160GLA0

UE2 gets 9 RBGs multiplexed with UE1

UE2

UE3 is scheduled in the remaining RBGs as single UE

UE3


RA41333EN160GLA0

113


LTE1787 TMP with 8Tx MU MIMO

TL15a

1 TTI UE1

UE2 UE1 TM8/9

RBG

1

Port 7

UE1

Port 8

2

3

4

UE3 UE2

UE4

5

6

UE5

UE6

UE5

pairing

Pairing process

UE2 TM8/9

9 common RBG

9 RBGs - UE 2

9 RBGs – UE1

RBG

1

Port 7

UE1

Port 8

UE2

2

UE2

3

4

5

6

UE3

UE3

UE5

UE6

UE2

UE4

UE5

6) 2) UE6 can’t be paired –UE1 conditions not UE2 is paired with UE3,are and it has 1) UE5 3) 5) UE1 UE3 works pared with with dual UE2 UE2 stream andand UE4 4) UE4 is paired with UE3 fulfilled one more unpaired RBG UEs can’t change the antenna port - all the time it is assigned to antenna port 7 or antenna port 8. M8010C70 M8010C66

115

RA41333EN160GLA0

TM9 dual-user single-stream mode TM8_DUAL_USER_SINGLE_BF_MODE


RA41333EN160GLA0

114


Capacity areas and cell resource measurements • Capacity areas • LTE QoS concept • Outer and Inner Loop Link Quality Control • Inner Loop and Outer Loop Link Adaptation DL • Inner Loop and Outer Loop Link Adaptation UL • Adaptive Transmission Bandwidth • UL and DL scheduling • MIMO • Power control • Cell throughput and availability • Cell resource Groups

116

RA41333EN160GLA0


RA41333EN160GLA0

115


Overview UL slow power control PPUSCH (i)  min {PCMAX ,10 log10 (M PUSCH (i))  PO_PUSCH ( j )   ( j )  PL  TF (i)  f (i)} dBm

Example: B Example: A PwR (dBm)

Closed loop PC: Based on exchange of feedback data and commands between UE and eNodeB

Open loop PC: Power

Min sensitivity

Max Path loss

Max Path loss

calculated by the UE based on path loss measurements "C1 calculation"

eNodeB PwR commands Max UE PwR needed no headroom

PC

time eNode B

117

RA41333EN160GLA0


PPUSCH (i) :PUSCH Power in subframe i PCMAX: max. allowed UE power (23 dBm for class 3) MPUSCH: number of scheduled RBs (The UE Tx. Power increases proportionally to # of PRBs) PO_PUSCH(j) = PO_NOMINAL_PUSCH(j) + PO_UE_PUSCH(j) PL: pathloss [dB] = referenceSignalPower – higher layer filtered RSRP TF (i) = 10 log 10 (2MPR Ks – 1) for Ks = 1.25 else 0, MPR = TBS/NRE, NRE : number of RE Ks defined by deltaMCS-Enabled, UE specific f(i): TPC (Closed Loop adjustment) Semi-persistant: j=0 / dynamic scheduling: j=1 PO_NOMINAL_PUSCH(0,1): cell specific (SysInfo) PO_UE_PUSCH(0,1): UE specific (RRC)  (0,1) = 0.0, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0 (partial PL compensation by open loop) Random access grant: j=2 PO_NOMINAL_PUSCH(2): PO_PRE + Preamble_Msg3 PO_UE_PUSCH(2) = 0  (2) = 1.0 (i.e. full PL compensation)

RA41333EN160GLA0

116


DL Static Power Setting

pMax

Total power going into PRBs = pMax - dlCellPerRed

LNCEL, 5/8/10/15/20/30/40/60/80W, -

• dMax = maximum output power for all DL channels • dlCellPwrRed = reduction of maximum output power • Available power evenly distributed over the PRBs (flat Power Spectral Density)

dlCellPwrRed LNCEL, 0..20 dB with step of 0.1 dB, 0 dB

PSD

PSD

= power per OFDM carrier

Number of OFDM carriers

PSD = (Max_TX_Pwr – CELL_PWR_RED) – 10*log10( 12*# PRBs)

Allocated DL PRBs

Frequency

DL Pilots 118

RA41333EN160GLA0

PSD = (Max_TX_Pwr – CELL_PWR_RED) – 10*log10( 12*# PRBs)

PDCCH

Time

PDSCH, PCH BCH, SCH © Nokia Solutions and Networks 2016

RA41333EN160GLA0

117


LTE1891 eNode B power saving - Micro DTX (FL16/TL16) • Micro-DTX is a feature that allows for POWER SAVINGS: • by switching OFF Power Amplifier(s) (PA) assigned to LTE Cell for the duration of single OFDM symbols • At times when the scheduler has determined that there is no DL signal to be sent over the air • Micro-DTX provides the MAXIMUM savings when: •

LTE cell is without any RRC connected UE

• Micro-DTX is NOT APPLICABLE when: •

MBMS (LTE 1117) is activated

•

RF sharing is activated

Counter name

Description

OFDM_SYM_MICRO_DTX_R_RFM (M8010C93)

The percentage of suppressed OFDM symbols with micro DTX to all transmitted symbols in RF module. The percentage is obtained from the average value of all samples. For each sample with a duration of 100 ms (1400 ODFM symbols), the percentage of suppressed symbols to all transmitted symbols is calculated. The average of these percentages is reported in a range of minutes, according to other RP1 transmissions. Trigger event: At the end of the measurement period, the average of all samples is calculated. Granularity: 100 ms KPI: LTE_5789a

#LTE Power and Quality DL

119

RA41333EN160GLA0


RA41333EN160GLA0

118


DL SINR measurement FDD only These measurements are updated for each UE every TTI by using the UE's most recent subband CQI to calculate Subband X downlink SINR post-compensation and the bin is incremented every sample period (5 seconds) with the respective number of calculated values where each one is within a dB range. PRB range of subband X according to frequency: 10 MHz refers to PRB#48 to PRB#49 15 MHz and 20 MHz refers to PRB#64 to PRB#71 M8031C0 .. M8031C19

SINR per subband distribution

DL_SINR_DIST_SB1_BIN0 .. DL_SINR_DIST_SB1_BIN19

M8031C240 … M8031C259

DL_SINR_DIST_SB13_BIN0 … DL_SINR_DIST_SB13_BIN19

M8031C260 .. M8031C279

DL_SINR_DIST_CW2_SB1_BIN0 .. DL_SINR_DIST_CW2_SB1_BIN19

M8031C500 … M8031C519

DL_SINR_DIST_CW2_SB13_BIN0 … DL_SINR_DIST_CW2_SB13_BIN19

M8031C520 .. M8031C539

DL_SINR_DIST_WB_BIN0 .. DL_SINR_DIST_WB_BIN19

M8031C540 .. M8031C559

DL_SINR_DIST_CW2_WB_BIN0 .. DL_SINR_DIST_CW2_WB_BIN19

16 14 12 10 8

28 to 30 dB 22 to 24 dB 18 to 20 dB 14 to 16 db 10 to 12 dB 6 to 8 dB 2 to 4 dB -2 to 0 dB -6 to -4 dB -10 to -8 dB

6 4

120

RA41333EN160GLA0

2 0


These measurements are updated for each UE every TTI by using the UE's most recent subband CQI to calculate Subband X downlink SINR post-compensation and the bin is incremented every sample period (5 seconds) with the respective number of calculated values where each one is within a dB range. PRB range of subband X according to frequency: 3 MHz and 5 MHz refers to PRB#0 to PRB#3 10 MHz refers to PRB#0 to PRB#5 15 MHz and 20 MHz refers to PRB#0 to PRB#7

RA41333EN160GLA0

119


DL SINR measurement FDD only Subbands mapping Subband 1 2 3 4 5 6 7 8 9 10 11 12 13

121

PRBs for 1.4 MHz N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

PRBs for 3 MHz 0 to 3 4 to 7 8 to 11 12 to 14 N/A N/A N/A N/A N/A N/A N/A N/A N/A

RA41333EN160GLA0

PRBs for 5 MHz 0 to 3 4 to 7 8 to 11 12 to 15 16 to 19 20 to 23 24 N/A N/A N/A N/A N/A N/A

PRBs for 10 MHz 0 to 5 6 to 11 12 to 17 18 to 23 24 to 29 30 to 35 36 to 41 42 to 47 48 to 49 N/A N/A N/A N/A

PRBs for 15 MHz 0 to 7 8 to 15 16 to 23 24 to 31 32 to 39 40 to 47 48 to 55 56 to 63 64 to 71 72 to 74 N/A N/A N/A

PRBs for 20 MHz 0 to 7 8 to 15 16 to 23 24 to 31 32 to 39 40 to 47 48 to 55 56 to 63 64 to 71 72 to 79 80 to 87 88 to 95 96 to 99


RA41333EN160GLA0

120


UL Power Control (SINR) • Main issue of PC is to keep a constant signal / interference + noise ratio at the receiver • In LTE interference is mainly inter-cell (power arriving from UEs which do not have an active radio link)

• Noise is thermal and hardware receiver noise, but also inter-modulation with other LTE carrier or even other RAT

Signal (S)

Low

Interference (I)

High

Noise (N)

122

RA41333EN160GLA0


Motivation

• In LTE orthogonal UL transmission,

i.e. near-far-problem much less severe than WCDMA • Nevertheless dynamic, slow open and closed loop PC • Path loss and shadowing compensation by open loop • Correction and adjustments of open loop inaccuracies by closed loop

Control of Power Spectral Density

• Power control does not control absolute UE power, but the Power Spectral Density (PSD) • PSDs at eNodeB from different users have to be close to each other so the receiver doesn’t work over a large range of powers • Different data rates mean different bandwidths so that the absolute power of the UE will also change. PC keeps PSD constant independently of the bandwidth

RA41333EN160GLA0

121


UL Power Control (Conventional and Fractional) • Conventional PC • Attempt to maintain a constant SINR at the receiver • UE increases the Tx power to fully compensate for increase of the path loss - Fractional PC scheme • Allow the received SINR to decrease as the path loss increases • UE Tx power increases with a reduced rate as the path loss increases. Increases of the path loss are only partially compensated • Partial instead of full compensation reduces inter-cell interference Improving air interface efficiency and increasing average cell throughput • Fractional PC specified by parameter ulpcAlpha ulpcAlpha

LNCEL, 0, 0.4..0.9 with step of 0.1, 1

UL SIN R

Conventional Power Control: α = 1

UL SIN R

UE Tx Power

UE Tx Power

If Path Loss increases by 10 dB the UE Tx power shall increase by 10 dB

123

RA41333EN160GLA0

Fractional Power Control: α < 1 If Path Loss increases by 10 dB the UE Tx power shall increase by α * 10 dB


RA41333EN160GLA0

122


UL Power and Quality Measurements • SINR

measurements M8005C90/C91/C92: Min/Max/Mean SINR on PUCCH M8005C93/C94/C95: Min/Max/Mean SINR on PUSCH M8005C96..C117: SINR histogram for PUCCH M8005C118..C139: SINR histogram for PUSCH The histograms work as follows 1.st counter: Number of measurements with SINR < -10 dB 2.nd counter: Number of measurements with SINR from -10 to -8 dB 3.rd counter: Number of measurements with SINR from -8 to -6 dB Continue with steps of 2 dB Last counter: Number of measurements with SINR > 30 dB

SINR on PUSCH/PUCCH 124

RA41333EN160GLA0

SNR > 30dB

28dB < SINR < 30 dB

Min

-6dB < SINR < -4dB

Mean

-8dB < SINR < -6dB

Max

-10 dB < SINR < -8 dB

-

SINR <10dB

-


LTE786 modifies the receiver at the eNodeB. The blanked PUCCH PRBs are not received, therefore they do not influence the received SINR. This means that the blanked resources do not contribute to the PUCCH RSSI nor SINR statistics, the measurements of the PUSCH RSSI and SINR are performed on the reduced amount of PRBs

RA41333EN160GLA0

123


UL Power and Quality Measurements • RSSI measurements M8005C0/C1/C2: Min/Max/Mean RSSI on PUCCH M8005C3/C4/C5: Min/Max/Mean RSSI on PUSCH M8005C6..C27: RSSI histogram for PUCCH M8005C28..C49: RSSI histogram for PUSCH The histograms work as follows 1.st counter: Number of measurements with RSSI < -120 dBm 2.nd counter: Number of measurements with RSSI from -120 to -118 dBm 3.rd counter: Number of measurements with RSSI from -118 to -116 dBm Continue with steps of 2 dB Last counter: Number of measurements with RSSI > -80 dBm

RSSI on PUSCH/PUCCH 125

RA41333EN160GLA0

RSSI> -80dB

-82dB< RSSI< -80dB

Min

-116dB < RSSI< -114dB

Mean

-118dB < RSSI< -116dB

Max -120 dB < RSSI< -118 dB

-

RSSI<120dB

-


LTE786 modifies the receiver at the eNodeB. The blanked PUCCH PRBs are not received, therefore they do not influence the received SINR. This means that the blanked resources do not contribute to the PUCCH RSSI nor SINR statistics, the measurements of the PUSCH RSSI and SINR are performed on the reduced amount of PRBs

RA41333EN160GLA0

124


UL Power and Quality Measurements • Interference to thermal noise measurements FDD only -

These measurements provide the 50 percentile value of the uplink Interference Power over Thermal (IoT) Noise Power for the PUSCH Resource Block (RB) with distribution bins 0-8.

-

M8005C317..C325: UL Interference over thermal noise for PUCSH To increase the granularity, the counter is reported in 10* dB units. Unit: 1.0E-1 dB

Bin 8

Bin 7

Bin 5

Bin6

Bin 4

Bin3

Bin2

Bin 1

Bin 0

dB

PRB Bandwidth

126

RA41333EN160GLA0


This measurements provide the 50 percentile value of the uplink Interference Power over Thermal (IoT) Noise Power for the PUSCH Resource Block (RB) distribution bin 0-8. To increase the granularity, the counter is reported in 10* dB units. To support the PUSCH RB distribution, there are 9 bins (bin0 to bin8). The PUSCH RBs would be evenly distributed (as much as possible) across the 9 bins with bin 0 and bin 8 having an equal number of PUSCH RBs.

RA41333EN160GLA0

125


UL Power and Quality Measurements • Distribution of SIR for PRBs of PUSCH FDD only -

These measurements provide the number of in-use PUSCH Resource Blocks (PRB) having an average Signal to Interference Ratio (SIR) value X

-

M8005C326..C335: UL Distribution of SIR for PRBs of PUSCH bins 0 - 9

Bin9 25dB > SIR

Bin8 22dB > SIR 25dB

Bin7 19dB > SIR < 22dB

Bin5 13dB > SIR < 16dB

Bin6 16dB >SIR < 19dB

Bin4 10dB > SIR <13dB

Bin3 7dB>SIR 10dB

Bin2 4dB >SIR <7dB

Bin1 1dB> SIR <4dB

Bin0 SIR <1dB

dB

PRB Bandwidth

127

RA41333EN160GLA0


This measurement is updated by calculating the SIR for each of the in-use PUSCH PRBs every sample period and incrementing the corresponding bin having an average SIR value X

RA41333EN160GLA0

126


UL Power and Quality Measurements • Received Total Wideband Power (RTWP) measurements FDD only -

These measurement are updated by accumulating the RTWP in milliwatts for Rx antenna 1-8 from every sample period and calculating the average at the end of the collection interval. To increase the granularity, the counter is reported in 10* dBm units.

-

M8005C306..C313: UL Interference over thermal noise for PUCSH To increase the granularity, the counter is reported in 10* dB units. Unit: 1.0E-1 dB

Max RTWP M8005C314 - This measurement provides the maximum Received Total Wideband Power (RTWP) at the end of collection interval.

Rx Antenna 8

Rx Antenna 7

Rx Antenna 5

Rx Antenna 6

Rx Antenna 4

Rx Antenna 3

Rx Antenna 2

Rx Antenna 1

dB

M8005C306..C313

128

RA41333EN160GLA0


This measurements provide the 50 percentile value of the uplink Interference Power over Thermal (IoT) Noise Power for the PUSCH Resource Block (RB) distribution bin 0-8. To increase the granularity, the counter is reported in 10* dB units. To support the PUSCH RB distribution, there are 9 bins (bin0 to bin8). The PUSCH RBs would be evenly distributed (as much as possible) across the 9 bins with bin 0 and bin 8 having an equal number of PUSCH RBs.

RA41333EN160GLA0

127


UL Power and Quality KPIs LTE_5541b E-UTRAN Average SINR for PUCCH

FDD

Shows the Signal to Interference and Noise Ratio (SINR) for physical UL control channel (PUCCH), measured in the eNodeB in dBm.

Logical formula AVG SINR PUCCH = average of measured SINR values for PUCCH

129

RA41333EN160GLA0



Avg ([M8005C92])

Avg ([SINR_PUCCH_AVG])


RA41333EN160GLA0

128


UL Power and Quality KPIs LTE_5544b E-UTRAN Average SINR for PUSCH Shows the Signal to Interference and Noise Ratio (SINR) for physical UL shared channel (PUSCH), measured in the eNodeB in dBm.

Logical formula AVG SINR PUSCH = average of measured SINR values for PUSCH

130

RA41333EN160GLA0



Avg ([M8005C95])

Avg [SINR_PUSCH_AVG])


RA41333EN160GLA0

129


UL Power and Quality KPIs 16.00

14.00

12.00

10.00

8.00

6.00

4.00

2.00

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

0.00

LTS_5541a Average SINR for PUCCH (dB) LTS_5544a Average SINR for PUSCH (dB)

131

RA41333EN160GLA0


RA41333EN160GLA0

130


UL Power and Quality KPIs LTE_5441b E-UTRAN Average RSSI for PUCCH Shows the average Received Signal Strength Indicator (RSSI) for physical UL control channel (PUCCH), measured in the eNodeB in dBm.

Logical formula AVG RSSI PUCCH = average of measured RSSI values for PUCCH

132

RA41333EN160GLA0



Avg ([M8005C2])

Avg ([RSSI_PUCCH_AVG])


RA41333EN160GLA0

131


UL Power and Quality KPIs LTE_5444b E-UTRAN Average RSSI for PUSCH Shows the average Received Signal Strength Indicator (RSSI) for physical UL shared channel (PUSCH), measured in the eNodeB in dBm.

Logical formula AVG RSSI PUSCH = average of measured RSSI values for PUSCH

133

RA41333EN160GLA0



Avg ([M8005C5])

Avg [RSSI_PUSCH_AVG])


RA41333EN160GLA0

132

RRM - Capacity Areas and Cell Resource Measurements

2…

1…

1…

1…

1…

1…

0…

0…

0…

3…

2…

2…

2…

2…

UL Power and Quality KPIs

0…

Slide 134

-80.00

-85.00

-90.00

-95.00

-100.00

-105.00

-110.00

LTE_5441a Average RSSI for PUCCH (dBm) LTS_5444a Average RSSI for PUSCH (dBm)

134

RA41333EN160GLA0


RA41333EN160GLA0

133


LTE1697 Timing Advance Histogram PM-counter Feature overview • The performance measurements introduced with this feature provide feedback about the distribution of subscribers in radial direction from the antenna based on instantaneous TA (timing alignment) values

d d

- TA values are sent to the UE to keep the uplink timing in sync with the downlink one - The TA value is provided by the Random Access response (initial TA) and all subsequent TA commands

d

d

• New counters are stored in new measurement group: -

M8029 – LTE MAC

-

Measurement interval parameter: PMRNL:mtMAC

d – range (in m) from the antenna

expectedCellSize LNCEL, 2.1,5,10,15,30,60,100km; 15km

135

RA41333EN160GLA0


Note

The histogram can reflect the actual range of the cell.

RA41333EN160GLA0

134


LTE1697 Timing Advance Histogram PM-counter Histogram configuration • Based on an operator-configured parameter (LNCEL:expectedCellSize) eNB is applying a pre-defined set of PM-counter bins (TAhistogram set) for mapping the measured instantaneous TA-value into a corresponding bin -

Each of the histogram PM-counters represents a bin for a certain distance range

#

The value of the parameter is always stored in additional counter TIMING_ADV_SET_INDEX (M8029C0) Counter bins

• Configuration of the 'expected cell size' can be performed -

Manually by the operator or

18000

-

Automatically during the system upgrade when doing the database conversion  Thereby a simple configuration rule is applied, which picks a proper expectedCellSize depending on the configured PRACH cyclic shift (prachCS) and high-speed flag (hsFlag).

16000

2.1 km range 5 km range

14000 Range [m]

Cell range

10 km range

12000

15 km range (def)

10000

8000 6000 4000

2000

expectedCellSize LNCEL, 2.1,5,10,15,30,60,100km; 15km

0

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Counter bin

136

RA41333EN160GLA0


RA41333EN160GLA0

135


LTE1697 Timing Advance Histogram PM-counter 2.1 km range

5 km range

10 km range

15 km range (def)

30 km range

60 km range

100 km range

Counter from

to

from

to

from

to

from

to

from

to

from

to

from

to

0

78

0

468

0

1006

0

1505

0

3003

0

6006

0

10000

M8029C2

78

156

468

1014

1006

2012

1505

3011

3003

6006

6006

12012

10000

19999

M8029C3

156

234

1014

1248

2012

2516

3011

3764

6006

7508

12012

15015

19999

24999

M8029C4

234

312

1248

1482

2516

3019

3764

4516

7508

9009

15015

18018

24999

29999

M8029C5

312

390

1482

1716

3019

3522

4516

5269

9009

10511

18018

21021

29999

34999

M8029C6

390

468

1716

2028

3522

4025

5269

6022

10511

12012

21021

24024

34999

39998

M8029C7

468

546

2028

2262

4025

4528

6022

6774

12012

13514

24024

27027

39998

44998

M8029C8

546

624

2262

2652

4528

5333

6774

7979

13514

15916

27027

31832

44998

52998

M8029C9

624

702

2652

3042

5333

6138

7979

9183

15916

18318

31832

36637

52998

60998

M8029C10

702

780

3042

3432

6138

6943

9183

10387

18318

20721

36637

41441

60998

68997

M8029C11

780

858

3432

3822

6943

7748

10387

11592

20721

23123

41441

46246

68997

76997

M8029C12

858

936

3822

3900

7748

8553

11592

12796

23123

25526

46246

51051

76997

84997

M8029C13

936

1014

3900

3978

8553

8653

12796

12946

25526

25826

51051

51652

84997

85997

M8029C14

1014

1092

3978

4056

8653

8754

12946

13097

25826

26126

51652

52252

85997

86997

M8029C15

1092

1170

4056

4134

8754

8855

13097

13248

26126

26426

52252

52853

86997

87996

M8029C16

1170

1248

4134

4212

8855

8955

13248

13398

26426

26727

52853

53453

87996

88996

M8029C17

1248

1326

4212

4290

8955

9056

13398

13549

26727

27027

53453

54054

88996

89996

M8029C18

1326

1404

4290

4368

9056

9156

13549

13699

27027

27327

54054

54655

89996

90996

M8029C19

1404

1482

4368

4446

9156

9257

13699

13850

27327

27628

54655

55255

90996

91996

M8029C20

1482

1560

4446

4524

9257

9358

13850

14000

27628

27928

55255

55856

91996

92996

M8029C21

1560

1638

4524

4602

9358

9458

14000

14151

27928

28228

55856

56456

92996

93996

M8029C1

137

RA41333EN160GLA0


RA41333EN160GLA0

136


LTE1697 Timing Advance Histogram PM-counter Bin mapping cont..... 2.1 km range

5 km range

10 km range

15 km range (def)

30 km range

60 km range

100 km range

Counter from

to

from

to

from

to

from

to

from

to

from

to

from

to

M8029C22

1638

1716

4602

4680

9458

9559

14151

14301

28228

28529

56456

57057

93996

94996

M8029C23

1716

1794

4680

4758

9559

9660

14301

14452

28529

28829

57057

57658

94996

95996

M8029C24

1794

1872

4758

4836

9660

9760

14452

14602

28829

29129

57658

58258

95996

96996

M8029C25

1872

1950

4836

4914

9760

9861

14602

14753

29129

29429

58258

58859

96996

97996

M8029C26

1950

2028

4914

4992

9861

9961

14753

14903

29429

29730

58859

59459

97996

98996

M8029C27

2028

2106

4992

5070

9961

10062

14903

15054

29730

30030

59459

60060

98996

99996

M8029C28

2106

2184

5070

5226

10062

10565

15054

15807

30030

31532

60060

63063

99996

104996

M8029C29

2184

2262

5226

5616

10565

11068

15807

16559

31532

33033

63063

66066

104996

109996

M8029C30

2262

infinite

5616

infinite

11068

infinite

16559

infinite

33033

infinite

66066

infinite

109996

infinite

138

RA41333EN160GLA0


RA41333EN160GLA0

137



139

RA41333EN160GLA0


RA41333EN160GLA0

138


Throughput and Latency Measurements • RLC level -

M8012C17: Received RLC data volume on UL in Kbyte

-

M8012C18: Transmitted RLC data volume on DL in Kbyte

• PDCP level -

M8012C21/C22/C23: Minimum/maximum/mean PDCP layer throughput on UL in Kbit/s

-

M8012C24/C25/C26: Minimum/maximum/mean PDCP layer throughput on DL in Kbit/s

-

M8001C3/C4/C5: Minimum/maximum/mean delay of a PDCP packet within eNodeB on UL in ms

-

M8001C0/C1/C2: Minimum/maximum/mean delay of a PDCP packet within eNodeB on DL in ms

PDCP Protocol and Cell Throughput related Trigger Counters

140

Counter Name

Expected Value

Test Value

PDCP_DATA_RATE_MEAN_DL

2048 (kbps)

2041

PDCP_DATA_RATE_MEAN_UL

2048 (kbps)

2040

PDCP_DATA_RATE_MIN_DL

2048 (kbps)

1943

PDCP_DATA_RATE_MIN_UL

2048 (kbps)

1599

PDCP_DATA_RATE_MAX_DL

2048 (kbps)

2187

PDCP_DATA_RATE_MAX_UL

2048 (kbps)

2476

RA41333EN160GLA0


Test Scenario : Perform Attach, then start download or upload by using Jperf with fixed 2 Mbps (1UE) then verify PM counters

RA41333EN160GLA0

139


Active cell time

RL50

New counters for improving U-plane throughput measurements in UL and DL Active cell time ACTIVE_TTI_UL (M8012C89) ACTIVE_TTI_DL (M8012C90)

Number of TTIs in UL with at least one UE scheduled to transmit user plane data (no C-plane/signaling/control information considered)

Trigger event: every TTI, in which at least one UE is scheduled to transmit/receive user plane data. Use case: This KPI enables new (more accurate) way of calculating the active cell throughput:

TTI with users scheduled to transmit User Plane data TTI without scheduled users No scheduled users in the cell

Data is transmitted

PDCP Layer Active Cell Throughput DL

8* (PDCP_SDU_VOL_DL) (LTE_5292d) = ACTIVE_TTI_DL

Time (ms) ACTIVE_TTI_* =

Count(

PDCP_SDU_VOL_ * =

Σ

) [#]

[bytes]

PDCP Layer Active Cell Throughput UL

8* (PDCP_SDU_VOL_UL) (LTE_5289d) = ACTIVE_TTI_UL

141

RA41333EN160GLA0


RA41333EN160GLA0

140


IP Protocol Statistics, Cell Throughput LTE_5350a…5358a

Averaged IP scheduled Throughput in DL, QCI1…QCI9

LTE_5359a..5367a

Averaged IP scheduled Throughput in UL, QCI1…QCI9

LTE_5503a…5511a

Averaged IP Throughput in DL, QCI1…QCI9

LTE_5512a…5520a

Averaged IP Throughput in UL, QCI1…QCI9

Includes TTIs where UE not scheduled, but has data in buffer Does not Include TTIs where UE not scheduled, but has data in buffer

Logical formula LTE5350a



IPSchedThrDLQCI1= IP Throughput Volume QCI1 in DL/ (IP Throughput Time QCI1 in DL)

sum([M8012C117])/sum([ M8012C118])

sum([IP_TPUT_VOL_DL_QCI_1])/ sum([IP_TPUT_TIME_DL_QCI_1])

Logical formula LTE5503a



IPThrDLQCI1= IP Throughput Volume QCI1 in DL/ (IP Throughput Scheduled Transmission Time QCI1 in DL)

sum([M8012C117])/sum([ M8012C165])

sum([IP_TPUT_VOL_DL_QCI_1])/ sum([IP_TPUT_NET_TIME_DL_QCI_1])

142

RA41333EN160GLA0


RA41333EN160GLA0

141


IP Protocol Statistics, Cell Throughput

M8012C89

Number of TTIs in UL with at least one UE scheduled to transmit user plane data

M8012C90

Number of TTIs in DL with at least one UE scheduled to receive user plane data.

M8012C91

IP Throughput volume in UL for QCI 1

M8012C92

IP Throughput time in UL for QCI 1

…..

M8012C107

IP Throughput volume in UL for QCI 9

M8012C108

IP Throughput time in UL for QCI 9

M8012C117

IP Throughput volume in DL for QCI 1

M8012C118

IP Throughput time in DL for QCI 1

… M8012C133

IP Throughput volume in DL for QCI 9

M8012C134

IP Throughput time in DL for QCI 9

M8012C156..164

IP Throughput net time in UL for QCI 1..9

M8012C165..173

IP Throughput net time in DL for QCI 1..9

143

RA41333EN160GLA0


RA41333EN160GLA0

142


PDCP Protocol Statistics, Cell Throughput LTE_5213a, M8012C19 LTE_5212a, M8012C20 M8012C21 M8012C22 LTE_5289d, M8012C23 M8012C24 M8012C25 LTE_5292d, M8012C26

PDCP SDU volume Uplink PDCP SDU volume Downlink Minimum PDCP Throughput UL Maximum PDCP Throughput UL PDCP Throughput UL Mean Minimum PDCP Throughput DL Maximum PDCP Throughput DL PDCP Throughput DL Mean

Logical formula LTE_5292d


8*sum([PDCP_SDU_VOL_DL])/sum([ACTIVE_ TTI_DL])

AVG DL PDCP CELL THP = average PDCP cell throughput DL

8*sum([M8012C20])/sum( [M8012C90])

8*sum([PDCP_SDU_VOL_DL])/sum([ACTIVE_ TTI_DL])

Logical formula LTE_5289d



AVG UL PDCP CELL THP = average PDCP cell throughput UL

8*sum([M8012C19]) /sum([M8012C90])

8*sum([PDCP_SDU_VOL_UL])/sum([ACTIVE_ TTI_DL])

Note: These PDCP Cell Throughput KPI‘s reflect Active Periods !

144

RA41333EN160GLA0


RA41333EN160GLA0

143


Throughput KPIs LTE_5283b E-UTRAN average RLC Layer Cell Throughput UL Shows the average RLC layer throughput per cell in uplink direction in Kbit/s.

Logical formula



AVG UL RLC CELL THP = (UL transmitted RLC PDU volume) * 8 / (MEASUREMENT_DURATION * 60)

Sum ([M8012C17]) * 8 / (sum(MEASUREMENT_ DURATION) * 60)

Sum ([RLC_PDU_VOL_TRANSMITTED]) * 8 / (sum (MEASUREMENT_DURATION) * 60)

Note: RLC Cell Throughput KPI‘s reflect Active and Inactive Periods !

145

RA41333EN160GLA0


RA41333EN160GLA0

144


Throughput KPIs LTE5284b E-UTRAN average RLC Layer Cell Throughput DL Shows the average RLC layer throughput per cell in downlink direction in Kbit/s.

Logical formula



AVG DL RLC CELL THP = (DL transmitted RLC PDU volume) * 8 / (MEASUREMENT_DURATION * 60)

Sum ([M8012C18]) * 8 / (sum(MEASUREMENT_ DURATION) * 60)

Sum ([RLC_PDU_VOL_TRANSMITTED]) * 8 / (sum (MEASUREMENT_DURATION) * 60)

146

RA41333EN160GLA0


RA41333EN160GLA0

145


E-UTRAN average RLC Layer Cell Throughput DL/UL 18,000.00 16,000.00 14,000.00 12,000.00 10,000.00 8,000.00

6,000.00 4,000.00 2,000.00

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

0.00

LTE_5283a Average RLC Layer Cell Throughput Uplink (kbit/s)

147

RA41333EN160GLA0


RA41333EN160GLA0

146


Latency KPIs LTE_5137a E-UTRAN Average Latency Uplink Shows the retention period (delay) of a PDCP SDU (UL) inside eNB Delay = time starting at the arrival of the PDCP SDU in the eNB and ending at the first transmission of a packet over S1 containing a segment of the SDU

Logical formula


LatencyAvgUL = PDCP SDU delay Avg ([M8001C5]) on UL DTCH Mean

148

RA41333EN160GLA0

Summarization formula (Abbreviation) Avg ([PDCP_SDU_DELAY_UL_DTCH_M EAN])


RA41333EN160GLA0

147


Latency KPIs The measurements providing the averaged PDCP SDU delay in DL per QCI given as averaged retention delay witthin the eNB related to since the PDCP SDU is received in eNB until its first part is sent via Uu interface

LTE_5471a – 5479a E-UTRAN Averaged PDCP SDU Delay in DL, QCI1- QCI9

Logical formula



PDCPSDUDelayDLQCI1= Average PDCP SDU delay in DL for QCI1

avg([]M8001C269] + [M8026C30])

avg([PDCP_RET_DL_DEL_MEAN_QCI _1] + [HARQ_DURATION_QCI1_AVG])

149

RA41333EN160GLA0


RA41333EN160GLA0

148


VOIP traffic KPIs LTE_5293c Average PDCP Layer Active Cell Throughput DL for QCI1 DRBs LTE_5294c Average PDCP Layer Active Cell Throughput UL for QCI1 DRBs



AVG DL PDCP CELL THP QCI1= average PDCP cell throughput DL for QCI1 DRBs

avg([M8012C143])

avg([PDCP_DATA_RATE_MEAN_DL_QCI_1])

AVG UL PDCP CELL THP QCI1= average PDCP cell throughput UL for QCI1 DRBs

avg([M8012C116])

avg([PDCP_DATA_RATE_MEAN_UL_QCI_1])

Logical formula

150

RA41333EN160GLA0


RA41333EN160GLA0

149



151

RA41333EN160GLA0


RA41333EN160GLA0

150


Cell Availability Measurements • Cell availability evaluated once approximately every 10s -

M8020C3: Number of evaluations the cell is available

-

M8020C4: Number of evaluations the cell is planned unavailable

-

M8020C5: Number of evaluations the cell in unplanned unavailable

-

M8020C6: Total number of evaluations

M8020C4 Cell unplanned unavailable (fault)



M8020C5 Cell planned unavailable (maintenance work)

M8020C3 Cell available

152

RA41333EN160GLA0




RA41333EN160GLA0

151


Cell Availability KPIs LTE_5750a E-UTRAN Cell Availability Ratio Shows the percentage of time the cell is available for services

Logical formula CELL AVR = (time of cell is available for services) / (total measured time) = (number of samples when cell is available) / (number of all samples)

153

RA41333EN160GLA0



Sum ([M8020C3]) / sum ([M8020C6] * 100

Sum ([SAMPLES_CELL_AV]) / Sum ([DENOM_CELL_AV]) * 100%


RA41333EN160GLA0

152


Cell Availability KPIs LTE_5751a E-UTRAN Planned Cell Unavailability Ratio Shows the percentage of time the cell is not available for services as planned by the operator (e.g. due to maintenance work)

Logical formula CELL PL UAVR = (time of cell is planned unavailable for services) / (total measured time) = (number of samples when cell is planned unavailable) / (number of all samples)

154

RA41333EN160GLA0



Sum ([M8020C4]) / sum ([M8020C6] * 100

Sum ([SAMPLES_CELL_PLAN_UNAV]) / Sum ([DENOM_CELL_AV]) * 100%


RA41333EN160GLA0

153


Cell Availability KPIs LTE_5752a E-UTRAN Unplanned Cell Unavailability Ratio Shows the percentage of time the cell is not available for services NOT planned by the operator (e.g. due to faults)

This KPI shows the ratio of services in a cell being unplanned unavailable for end-users.

Logical formula CELL UPL UAVR = (time of cell is unplanned unavailable for services) / (total measured time) = (number of samples when cell is unplanned unavailable) / (number of all samples)

155

RA41333EN160GLA0



Sum ([M8020C5]) / sum ([M8020C6] * 100

Sum ([SAMPLES_CELL_UNPLAN_UNAV ]) / Sum ([DENOM_CELL_AV]) * 100%


RA41333EN160GLA0

154


Cell Availability KPIs 100.00

98.00

96.00

94.00

92.00

LTE_5750a Cell Availability Ratio (%)

156

RA41333EN160GLA0

21.06.2011

19.06.2011

17.06.2011

15.06.2011

13.06.2011

11.06.2011

09.06.2011

06.06.2011

04.06.2011

02.06.2011

31.05.2011

29.05.2011

27.05.2011

25.05.2011

23.05.2011

90.00

LTE_5239a Cell Availability excluding Blocked by User (%)


RA41333EN160GLA0

155


eNB Load (M8018) LTE eNB Load measurement (M8018) measures load and capacity relevant internal indicators per eNB. M8018C0 / Active UE per eNB average Updated: The arithmetical average of samples taken from the number of UEs, having one SRB and at least one DRB established. The length of the sampling interval is 4 seconds. M8018C1 / Active UE per eNB max Updated: After change of number of active UEs

157

RA41333EN160GLA0


M8018C0 The average number of active UEs per eNB. A UE is defined as "active" if at least a single non-GBR DRB has been succesfully configured for it. Updated: The arithmetical average of samples taken from the number of UEs, having one SRB and at least one DRB established M8018C1 Description: The maximum number of active UEs per eNB. A UE is defined as "active" if at least a single non-GBR DRB has been successfully configured for it. Updated: The maximum value of the samples of UEs having one SRB and at least one DRB established during a measurement period.

RA41333EN160GLA0

156



158

RA41333EN160GLA0


RA41333EN160GLA0

157


LTE1382: Cell resource groups RAN-based Network Sharing solutions for LTE • LTE1382 Cell Resource Groups is a new feature which improves Network Sharing concept in LTE • LTE Radio Access Network*, can be shared with the help of two network sharing features: - MORAN (Multi-Operator RAN) which is NSN proprietary solution • each Operator has its own Core Network nodes while eNB equipment is shared between the Operators • moreover each Operator has its own dedicated cells - MOCN (Multi-Operator Core Network) which is standardized by 3GPP • similar to MORAN, each Operator has its own Core Network nodes while eNB equipment is shared • in MOCN Operators shares also eNB cells i.e. cell resources are available for users of both (or more) Operators

159

RA41333EN160GLA0


(*): 3GPP specifies also other network sharing solutions for LTE, namely National Roaming and Gateway Core Network however those are Core based solutions which are out of scope of this presentation

RA41333EN160GLA0

158


LTE1382: Cell resource groups •

Provides PDSCH, PUSCH resources reservation at cell level,

•

Up to 4 Cell Resource Groups (CRG) → each CRG represents one or more PLMN Ids

–

A TTI is assigned to each of the Cell Resource Groups in WRR in accordance with weighting factor



plmnGroupShare

•

Introduces also reservation of Admission Control resources:

–

Admission Control resources in the cell are distributed between CRGs (based on plmnGroupShare)

For example with 50/50 split: • each Operator has assured AC capacity in terms of number of Active UEs, DRBs, QCI1 users • each Operator can use all PRBs in every second TTI

160

RA41333EN160GLA0


•Provides PDSCH, PUSCH resources reservation at cell level, among up to 4 Cell Resource Groups (CRG) → each CRG represents one or more PLMN Ids – reservation is realized by assigning a complete TTI to one of the Cell Resource Groups – selection of a CRG for scheduling in a TTI is done by Weighted Round Robin algorithm according to a configured share  split of resources is configurable by plmnGroupShare parameter

•Introduces also reservation of Admission Control resources: – Admission Control resources in the cell are distributed between the Operators according to configurable share (also plmnGroupShare is used)

RA41333EN160GLA0

159


LTE1382: Cell resource groups Performance Management

Counter

name

M8011C70

TTIs used in UL by cell resource group 1

M8011C71


M8011C72


M8011C73


M8011C74

Number of available TTIs for PUSCH

M8011C75

TTIs used in DL by cell resource group 1

M8011C76


M8011C77


M8011C78


M8011C79

Number of available TTIs for PDSCH

161

RA41333EN160GLA0


RA41333EN160GLA0

160


LTE1382: Cell resource groups Performance Management

KPI

name

formula

LTE_5411a

E-UTRAN Cell Resource Group 1 Utilization Ratio in DL

100*sum([M8011C75])/ sum([M8011C79])

LTE_5412a E-UTRAN Cell Resource Group 2 Utilization Ratio in DL

100*sum([M8011C76])/ sum([M8011C79])


100*sum([M8011C77])/ sum([M8011C79])


100*sum([M8011C78])/ sum([M8011C79])

LTE_5415a E-UTRAN Cell Resource Group 1 Utilization Ratio in UL

100*sum([M8011C70])/ sum([M8011C74])


100*sum([M8011C71])/ sum([M8011C74])


100*sum([M8011C72])/ sum([M8011C74])


100*sum([M8011C73])/ sum([M8011C74])

162

RA41333EN160GLA0


RA41333EN160GLA0

161


RA41333EN160GLA0


RA41333EN160GLA0

162

03 RA41333EN160GLA0 LTE Capacity Areas

Recommend Documents