210185358 LTE Training RF Design Ed4

From LTE basics to 9155 LTE RF Design

September 2009

LTE Basics OFDM Fundamentals

2 | Presentation Title | Month 2008

All Rights Reserved © Alcatel-Lucent 2008, XXXXX



Basic of OFDM



Basic of OFDM Waveform



Basic of OFDM Sending modulation symbol in parallel



Basic of OFDM Symbol extract



Basic of OFDM



Basic of OFDM Orthogonality lost



Basic of OFDM Doppler & frequency offset effects



Basic of OFDM Multi-path effect



Basic of OFDM Multi-path effect



Basic of OFDM CP length



Basic of OFDM OFDM scalable



Basic of OFDM Full Tx/Rx chain



LTE Basics DOWNLINK STRUCTURE



DL Physical Channels



DL Channels Mapping



LTE Downlink: Frame Format, Channel Structure & Terminology



LTE Downlink: Number of Resource Blocks & Numerology



Downlink common Reference Signal structure

 Reference signal symbol distribution sequence over 12 subcarriers x 14 OFDM symbols.  The Reference signal sequence is correlated to Cell ID. 21 | Presentation Title | Month 2008


Downlink common Reference Signal structure per number of antenna port



PBCH, SCH Time and frequency location



Basic of cell search



Primary BCH & Dynamic BCH



Primary BCH & Dynamic BCH



PCFICH & PHICH



PDCCH



PDCCH: DCI formats carried DCI includes resource assignments and other control information



Downlink Shared Channel (DL-SCH)



DL Power settings PDCCH

PBCH

Based o the simus done by R&D and also on first trials results the DL power settings is detailed in the slides below



DL Power settings LA 0.x



DL Power settings LA 1.0 RRH 30W



DL Power settings LA 1.0 RRH 40W



LTE Basics UPLINK STRUCTURE



UL Physical Channels



UL Channels Mapping



SC-FDMA principle



SC-FDMA principle



SC-FDMA Tx/Rx chain



LTE Uplink: Number of Resource Blocks & Numerology



Demodulation Reference Signal & Sounding Reference Signal



Demodulation Reference Signal & Sounding Reference Signal



PUCCH



PUCCH



PUCCH



PRACH



Radom Access procedures



LTE Basics UL Power Control



IoT Control Mechanism (Inter-cell Power Control)  Setting of Target_SINR_dB determines the IoT operating point  Especially in a reuse-1 deployment, it is critical to manage the uplink interference level  In LTE, e-NBs can send uplink overload indications to neighbor e-NBs via the X2 interface  Power control parameters (i.e. Target SINR) can be adapted based on overload indicators  Allows control of the IoT level to ensure coverage and system stability Overload indicator (X-2 interface) PC params

PC params Measure Interference, emit overload indicator


interference


Based on overload indicator from neighbor cell, adapt PC params

Fractional Power Control

 While using the same target SINR for each user results in very good fairness (as far as power allocation is concerned), it also results in poor spectral efficiency  An improved power control scheme called Fractional Power Control adjusts the target SINR in relation to the UE’s path loss to its serving sector  UE_TxPSD_dBm = a x PL_dB + Nominal_Target_SINR_dB + UL_Interference_dBm  a is called the fractional compensation factor, and is sent via cell broadcast; 0 < a<1 Target_SINR_dB = Nominal_Target_SINR_dB - (1-a) x PL_dB Target SINR increases with decreasing path loss Flexible trade-off between cell edge rate and average spectral efficiency 51 | Presentation Title | Month 2008


Target SINR

Improved Power Control Based on Neighbor Cell Path Loss

 Path loss to the serving cell is not indicative of the amount of interference a user will generate to neighboring sectors  An improved power control scheme adjusts the target SINR in relation to DPL_dB = PL_strongestNeighborCell_dB – PL_servingCell_dB  UE_TxPSD_dBm = PL_dB + Nominal_Target_SINR_dB + (1-b) x DPL_dB + UL_Interference_dBm  (1-b) x DPL_dB is sent to each UE via higher layer (RRC) signaling

Target_SINR_dB = Nominal_Target_SINR_dB + (1-b) x DPL_dB

Target SINR increases with increasing “radio position”



Target SINR

LTE Basics Scheduler



Scheduler



UL Scheduling mechanism



DL Scheduling mechanism



Channel Quality Indicator, Pre-coding Matrix Indicator, Rank Indicator



Scheduler weighted proportional fair



Scheduler proportional fair principles












Frequency Non-Selective Scheme 6

Single priority metric formed and used in the first stage of the MPE algorithm

5

4

3

Priority Metric

Then MPE algorithm continues as in FSS scheme

2

1

UE 1 UE 2 UE 3

0

UE 3 UE 2 UE 1 1

2

3

4

5

6

7

8

9

Resource Unit Index

 The SRS SYNC SINR is a scalar quantity per user that is formed by averaging the SRS SINR across PRBs and then filtered in time; used to form a single priority metric, which is replicated and used for all PRBs  To support a large number of UEs, the SRS period needs to be reduced given the multiplexing capabilities (max of 8 UEs per SRS transmission per frequency comb)

 The regular MPE algorithm as in the FSS algorithm is then utilized, which minimizes testing/verification to just the new code introduced  Currently also investigating an intermediate solution where the resolution of the frequency selective scheduler is reduced by a certain factor in order to retain some frequency selectivenessin the scheduling while reducing complexity (study in progress) 63 | Presentation Title | Month 2008


Frequency Re-use strategies

Frequency re-use1


Fractional Frequency re-use


Frequency Re-use strategies Soft Frequency re-use or dynamic frequency re-use



LTE Basics Link adaptation



DL MCS table



UL MCS table



LTE Basics Multi Antenna Technology Roadmap



MIMO Configuration



Antennas Configuration



Antennas Configuration



Spatial Multiplexing



LA1.0 Scheme supported



Scheme supported after LA1.0



LTE Link Budgets Uplink Link Budget Considerations

Uplink Link Budget Main Principles

Link Budget is performed for one mobile located at cell edge (for each service) transmitting at max power The IoT (Interference over Thermal Noise) experienced by this user on the UL depends on the frequency reuse scheme and the service data rate and corresponding SINR that is guaranteed for cell edge users

UPLINK Analysis is an MAPL analysis

MAPL Max UE transmit Power Required Received Signal cell radius



Uplink Link Budget Main Principles UL link budget elaborated for user of service k at cell edge transmitting at maximum power

Uplink Path

Transmit Power

UE Transmit power (23dBm)

Losses and Margins

Feeder losses Penetration Loss (outdoor/indoor) Shadowing Margin

Gains

eNode-B Antenna Gain UE Antenna Gain

Receiver Sensitivity

Derived from SINR performances

Handoff Gain Body Loss



Interference

= MAPL

Interference Margin

Maximum Allowable Path Loss

Uplink Link Budget Rationale Behind LKB Formulation UL Rates 128kbps 256kbps 512kbps

RangeUL_Guar_Serv

Link budgets are formulated for one service that is to be guaranteed at cell edge (RangeUL_Guar_Serv) For more limiting service rates link budgets are formulated under the assumption they are not guaranteed at cell edge but at a reduced coverage footprint



Uplink Link Budget Example for one service

Dense Urban (2.6GHz) Required Data Rate No. Resource Blocks Required MCS Used Bandwidth Target C/I eNode-B Noise Figure eNode-B Sensitivity Antenna Gain Cable & Connector Losses Body Losses Additional UL Losses Cell area coverage probability Overall standard deviation Shadowing Margin Handoff Gain Fast Fading Margin Penetration Margin Fixed IoT UE Antenna Gain UE Max Transmit Power MAPL UL Cell Range 80 | Presentation Title | Month 2008

PS 128 128 kbps 3 RB MCS 8 540 kHz -3.0 dB 2.5 dB -117.2 dBm 18.0 dBi 0.5 dB 0 dB 0 dB 95% 8.0 dB 8.6 dB 3.6 dB 0 dB 21 dB 3.0 dB 0 dBi 23.0 dBm 128.7 dB 0.46 km

No. Resource Blocks to Reach Data Rate Optimal Modulation & Coding Scheme (MCS) Signal to Interference Ratio per Resource Block Noise Figure of the eNode-B is supplier dependent Based on SINR, Noise Figure, Thermal Noise, Bandwidth Used


Uplink Link Budget Receiver Sensitivity

eNode-B Receiver Sensitivity Minimum required signal level to reach a given quality (SINR target) when facing only thermal noise SensitivitydBm

= SINRdB + 10.log10(F.Nth.NRB.WRB)

 Where:  F: eNode-B Noise figure in dB

Service dependent

 Nth: Thermal noise density, 10log(Nth) =-174 dBm/Hz  SINRdB: Signal to Interference ratio per Resource Block  NRB: Number of resource blocks (RB) required to reach a given data rate  WRB: Bandwidth of one Resource Block – One Resource Block is composed of 12 subcarriers, each of a 15kHz bandwidth – so WRB = 180kHz.\ 81 | Presentation Title | Month 2008


Uplink Link Budget SINR Performances - Overview

SINR Target depends on:  eNode-B equipment performance

 Radio conditions (multipath fading profile, mobile speed)  Receive diversity (2-way by default or optional 4-way)  Targeted data rate and quality of service

 The Modulation and Coding Scheme (MCS)  Max allowed number of HARQ transmissions (Maximum of 4 on UL)  HARQ Operating Point – 1% Post HARQ BLER target considered by default Derived from link level simulations or better by equipment measurements (lab or on-field measurements)



Uplink Link Budget SINR Performances - Channel Model

In reality, a mix of multipath conditions exist across a typical cell  For coverage assessment, the worst case model should be considered

 ITU VehA multipath channel model are considered a good compromise For LTE some evolved multipath channel models have been defined such as EVA5Hz or EPA5Hz  These are an extension of the VehA and PedA models used in UMTS to make them more suitable for the wider bandwidths encountered with LTE, e.g. >5MHz  Main difference lies in the definition of a Doppler frequency instead of a speed, making the model useable for different frequency bands  All SINR performances used in the link budget are for all EVehA3 and EVehA50 channel models



Uplink Link Budget SINR Performances - Link Level Results for 10MHz Bandwidth (50 RB) LTE UL Throughput v.s. SNR, max 4HARQ Tx, EPedB-3km

40000

MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS MCS

35000

Throughput (kbps)

30000

25000

20000

= = = = = = = = = = = = = = =

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

MCS = 1 MCS = 3 MCS = 5 MCS = 7 MCS = 9 MCS = 11 MCS = 13 MCS = 15 MCS = 17 MCS = 19 MCS = 21 MCS = 23 MCS = 25 MCS = 27 T'put (kbps)

15000

10000

5000

0 -10

-5

0

5

10

15

SINR (dB)



20

25

30

Uplink Link Budget SINR Performances - Selection of Optimal SINR Figures

There are a number of possible solutions that can be used to provide a given throughput – solutions comprise a combination of:  Modulation & Coding Scheme (MCS)  Number of Resource Blocks (RB) Optimization Objective:  Select # RB’s and MCS so as to maximize the receiver sensitivity and thus the link budget  While at the same time respecting the selected HARQ operating point (1% post HARQ BLER objective)



Uplink Link Budget SINR Performances - Summary for UL 10MHz Bandwidth (1x2 RxDiv)

Performance figures for typical UL link budget rates  Number of RB’s

 SINR (include margins)  MCS, TBS and # HARQ Transmissions Service

VoIP

PS 64

PS 128

PS 256

PS 384

PS 512

PS 768

Bit Rate

12.2

64

128

256

384

512

768

1000

2000

MCS

6

6

8

10

10

10

10

10

10

TBS

328

176

392

872

1384

1736

2792

3496

6968

QPSK

QPSK

QPSK

QPSK

QPSK

QPSK

QPSK

QPSK

QPSK

Post HARQ BLER

1%

1%

1%

1%

1%

1%

1%

1%

1%

Required # of RB

1

2

3

5

8

10

16

20

40

-3.6 dB

-3.0 dB

-2.4 dB

-2.9 dB

-3.1 dB

-3.4 dB

-2.9 dB

-3.3 dB

Modulation

SINR (EVehA 3km/h) -3.7 dB

PS 1000 PS 2000

Rx Sensitivity -123 dBm -120 dBm -117 dBm -114 dBm -113 dBm -112 dBm -110 dBm -109 dBm -106 dBm



Uplink Link Budget SINR Peformances - MCS and TBS Tables

Some Background Info  Modulation & Coding Scheme (MCS)  This determines the Modulation Order which in turn determines the TBS Index

 Number of Resource Blocks  For a given MCS the Transport Block Size (TBS) is given different numbers of resource blocks TBS Table

MCS Table NPRB MCS Index, IMCS

Modulation Order, QM

TBS Index, ITBS

0 1 2 3 …

QPSK QPSK QPSK QPSK QPSK

0 1 2 3 4


ITBS 0 1 2 3 4 5 6 …

1 16 24 32 40 56 72 328 104


2 32 56 72 104 120 144 176 224

3 56 88 144 176 208 224 256 328

4 88 144 176 208 256 328 392 472

… 120 176 208 256 328 424 504 584

Uplink Link Budget Implementation Margins

SINR performances from link level simulations assume ideal scheduling and link adaptation – reality will not be as good …  For example in the downlink, we consider: Error free CQI feedback, Perfect PDCCHPCFICH decoding, CQI feedback rate 1/20ms, etc.

To account for such ideal assumptions there are currently two key elements to the margins incorporated into in SINR performances used in UL budgets today:

 Implementation margin to account for the assumptions implicit in the link level simulations used to derive the SINR performances  Currently considered to be ~1dB  No variability is assumed for different environments or UE mobility conditions

 Will be tuned based on SINR measurements (not yet performed)

 ACK/NACK margin to account for the puncturing of ACK/NACK onto the PUSCH  A 1dB margin is applied for VoIP services and 0.5dB for higher data throughputs



Uplink Link Budget Consideration of Explicit Diversity Gains

The SINR performance figures considered by Alcatel-Lucent in UL and DL link budgets are based on link level simulations that already account for the corresponding transmit and receive diversity gains, i.e.  UL: default 1x2 Rx Diversity  2RxDiv gain accounted for in the SINR figures  To account for 4RxDiv on the UL an additional 2-3dB gain is considered on the 2RxDiv SINR figures

 DL: default 2x2 Tx Diversity  SFBC pre-coding gains + 2RxDiv gain at the UE are accounted for in the SINR figures  Note that an additional power combining gain is considered at the transmit side, i.e. for a 2 x 40W TxDiv configuration a 80W transmit power is applied in DL link budgets



INTERNAL NOTE – Noise Figure The Noise Figure of the eNode-B is supplier dependent  Typically the Noise Figures of e-NodeBs range between 2 to 3dB Typical RRH Noise Figures for ALU product (June 2009)

RRH Type RRH2x (lower 700) 900 MC-TRX (1800) MC-RRH (1800) AWS RRH2x (2600) TRDU (2600)


Typical Noise Figure 2.2dB TBD - 2.5dB (assumed) 3 dB 2.5 dB TBD – 2.5dB (assumed) 2.6 dB 2.6 dB


Uplink Link Budget Exercise

Compute eNode-B sensitivity in VehA 3km/h  for VoIP 12.2kbps @ 1% Post-HARQ BLER

 For PS 384kbps @ 1% Post-HARQ BLER

Alcatel-Lucent equipment:  Typical eNode-B Noise Figure: 2.5dB  SINR figures: -3.7 dB for VoIP 12.2, -3.3dB for PS384

ANSWER: Sensitivity: -122.6 dBm for speech, -113.2 dBm for PS384






Typical gain of Tri-sectored antenna, depends on frequency band Depends on feeder type, length and frequency band 3dB body loss when speech usage (UE near head), 0dB body loss when data usage

0dBi by default Depends on UE Power Class


Uplink Link Budget UE Characteristics

LTE UE Max Transmit Power  Depends on the power class of the UE

 Only one power class is defined in 3GPP TS 36.101: 23dBm output power is considered with a 0 dBi antenna gain; ± 2dB tolerance in the standard WCDMA UE Max Transmit Power  Multiple power classes were defined in 3GPP TS 25.101, the most prevalent WCDMA UE’s today are considered to be class 3 (24dBm +1/-3dB) The corresponding tolerance ranges for both WCDMA and LTE terminals are in fact the same:  4dB range 21-25dBm  While the nominal Tx powers differ by 1dB  Currently consider 23dBm in UL LTE link budgets 93 | Presentation Title | Month 2008





Shadowing margin due to shadowing standard deviation Handoff gain Depends on depth of coverage (e.g. deep indoor, indoor daylight, outdoor). Also accounts for the indoor shadowing margin


Uplink Link Budget Shadowing Margin

Shadowing Margin:  Slow fading signal level variations due to obstacles

 Modelled (in dB) as a Gaussian variable with zero-mean and standard deviation depending on the environment, typically 6 to 8dB  The shadowing standard deviation can include the variability associated with the indoor penetration. However, it is recommended to consider this as part of the penetration margin Impact on link budget :  Take a margin to ensure the received signal is well received (above required sensitivity) with a given probability  Typically 95% in Dense Urban, Urban and Suburban and 90% in Rural  Computation as for UMTS and CDMA.



Uplink Link Budget Handoff Gain Unlike UMTS/WCDMA or CDMA, there is no soft-handoff functionality for LTE

 No soft-handoff gain considered for LTE Far too pessimistic to only consider the shadowing margin computed with one cell unless considering an isolated cell A mobile at the cell edge can still handover to a neighbor cell with more favorable

shadowing, i.e. a lower path loss  consider a Handoff Gain (or best server selection gain) Reference article: Analysis of fade margins for soft and hard handoffs, Rege, K.M.; Nanda, S.; Weaver, C.F.; Peng, W.-C., PIMRC 95 INTERNAL NOTE: This hard handoff gain can be considered for any system without soft handoff. So this is the case for GSM. However no gain is typically applied in GSM. For LTE the sampling frequency for handoff decisions as well as the handoff speed itself is much faster than GSM  this leads to an LTE handoff gain not much less than that considered for WCDMA.



Uplink Link Budget Handoff Gain - Example Antenna Height

30 m

K2 Propagation Model

35.2

Shadowing Correlation

0.5

Hysteresis

2 dB

HO sampling time # of samples to decide HO

20 msec 4 samples

Correlation distance

50 m

Cell Range

100%

Typical for Suburban Incar & Rural

Typical for Dense Urban, Urban and Suburban Indoor

Shadowing Standard Deviation

6 dB

6 dB

7 dB

7 dB

8 dB

8 dB

10 dB

10 dB

Cell Area Coverage Probability

90%

95%

90%

95%

90%

95%

90%

95%

Cell Edge Coverage Probability

71%

84%

73%

85%

75%

86%

78%

88%

Handoff Hysteresis

2 dB

2 dB

2 dB

2 dB

2 dB

2 dB

2 dB

2 dB

Shadowing Margin (no SHO gain)

3.3 dB

5.9 dB

4.3 dB

7.2 dB

5.4 dB

8.7 dB

7.7 dB

11.7 dB

SHO Gain

2.7 dB

2.8 dB

3.1 dB

3.4 dB

3.6 dB

3.9 dB

4.7 dB

5.0 dB

3 km/h - HHO Gain

2.3 dB

2.5 dB

2.8 dB

3.1 dB

3.4 dB

3.6 dB

4.4 dB

4.8 dB

50 km/h - HHO Gain

2.1 dB

2.2 dB

2.6 dB

2.8 dB

3.1 dB

3.3 dB

4.1 dB

4.4 dB

100 km/h - HHO Gain

2.0 dB

2.0 dB

2.4 dB

2.6 dB

2.8 dB

3.0 dB

3.7 dB

4.0 dB

Reference article: Analysis of fade margins for soft and hard handoffs, Rege, K.M.; Nanda, S.; Weaver, C.F.; Peng, W.-C., PIMRC 95 97 | Presentation Title | Month 2008


Uplink Link Budget Handoff Gain - Example

Note that the full Handoff Gain is only applicable for UE’s located at the cell edge  where we consider one rate guaranteed at the cell edge and others guaranteed within that coverage footprint, the other services will not take 4.0 dB benefit of the full handoff gain Handoff Gain

3.5 dB 3.0 dB 2.5 dB 2.0 dB 1.5 dB 1.0 dB 0.5 dB 128kbps 256kbps 512kbps

UL Rates

0.0 dB 0%

20%

40%

60%

% of Cell Range

Dense Urban, Sigma = 8dB, 95% coverage reliability, 3km/h mobility 98 | Presentation Title | Month 2008


80%

100%

Uplink Link Budget Penetration Margin

The penetration losses characterize the level of indoor coverage targeted by the operator (deep indoor, indoor daylight, window, incar, outdoor, etc)  Highly dependent on the wall materials and number of walls/windows to be penetrated It is recommended to consider the penetration margin as a single “worst case” margin as the shadowing standard deviation doesn’t include the indoor penetration variability Typical Penetration Losses at 2GHz Environment

Penetration Margin (dB)

Dense Urban – Deep Indoor

20

Urban - Indoor

17

Suburban - Indoor

14

Rural – Incar

8



INTERNAL NOTE – Penetration Losses For 700/850/900MHz, lower penetration losses can be considered  Note that the frequency dependency of the penetration losses is very materialdependent  Typically, we can assume 2dB lower penetration margins compared to those at 2GHz For 2.6GHz, higher penetration losses could be considered  Note that the frequency dependency of the penetration losses is very materialdependent  Typically, we can assume 2dB higher penetration margins compared to those at 2GHz






This sensitivity is calculated for noise only. A margin must be considered for the interference above noise: Interference Margin

Interference Margin or IoT


Uplink Link Budget Interference Margin

Sensitivity figures typical consider only thermal noise, the real interference, Ij, must also be considered (not only the thermal noise) Received Power, C jdBm  Sensitivit y dBm  InterferenceMargindB  Ij   InterferenceM argindB  10log   Nth W 

Interference margin or IoT (Interference over Thermal Noise)  A reuse of 1 is typical (option to use schemes such as soft fractional reuse or interference coordination)

 The IoT operating point can be set to achieve a minimum data rate at cell edge and/or to match incumbent technology coverage



Uplink Link Budget WCDMA Noise Rise - What’s Different Between LTE and WCDMA?

By definition, Cell Load and Total Interference rise (“Noise Rise”) are linked:

I  i tot _ dB  10log  total   10log 1  xUL  No W   where Itotal is the total received power at the node B (including the useful signal) Differences with LTE  Interference from adjacent cells only for LTE (no intracell interference)  Max WCDMA cell load is dependent on power control stability  No concept of cell load for LTE

Noise Rise (dB)

30 25 20 15 50% cell load 3dB Noise Rise

10 5 0 0

10

20

30

40

50

60

70

Cell Load (%)



80

90

100

LTE IoT What Determines the IoT for LTE?

 The average IoT is dependent upon the targeted cell edge data rate (SINR)  The higher the cell edge SINR target, the higher the average IoT

 Ultimately there is a point at which the increased IoT can not be sustained with the corresponding SINR Based on system level simulations: 400

9

Average Throughput

Avg and 5% UE Throughput (kbps)

8

Average IoT (dB)

7 6 5 4 3 2 1 0

Cell Edge Throughput

300

200

100

0

-7

-6

-5

-4

-3

-2

-1

0

1

2

0

1

3

4

5

Mean IoT (dB)

Cell Edge SINR Target, TSINR (dB)


2


6

7

8

9

LTE IoT What Determines the IoT for LTE?

For LTE the IoT can be expressed as: IoT = 1 / (1 - RBLoad x FAvg x TSINR)

Where  RBLoad = Average % loading of the resource blocks of adjacent cells  Under full loading this can be considered to be 100%

 FAvg = The average ratio between extracell interference and useful signal received at the eNode-B  Based on system level simulations the typical value of FAvg for UL fractional power control is ~0.8 – this is quite comparable to that used for WCDMA

 TSINR = SINR target at the cell edge



LTE IoT The IoT for Targeted LTE Cell Edge Rates? 10 MHz bandwidth - 2RxDiv - Product Release LAx VoIP AMR 12.2 with TTI Bundling

Modulation Coding Rate Max # HARQ Tx Post-HARQ BLER Required # of RB

12.2 QPSK 0.31 4 1% 1

VehA 3km/h VehA 50km/h

-3.7 -2.1 RBLoad FAvg

VehA 3km/h VehA 50km/h

1.8 dB 3.0 dB

PS 8

PS 64

PS 128

PS256

PS 500

PS 1Mbps PS 2Mbps

8 64 128 256 384 500 QPSK QPSK QPSK QPSK QPSK QPSK 0.14 0.35 0.48 0.62 0.61 0.61 4 4 4 4 4 4 1% 1% 1% 1% 1% 1% 1 2 3 5 8 10 SNR Figures @ 2.6GHz (including implementation and ACK/NACK -3.38 -3.4 -2.9 -2.7 -3.3 -3 -3.8 -2.8 -2.6 -2.1 -2.5 -2.5

1000 QPSK 0.62 4 1% 21 margins) -3.3 -2.7

2000 QPSK 0.61 4 1% 41 -3.4 -2.9

100%

IoT = 1 / (1-RBLoad.FAvg.TSINR)

0.8 2.0 dB 1.8 dB

2.0 dB 2.4 dB

2.3 dB 2.5 dB

2.4 dB 3.0 dB

IoT for 100% RB Loading Ranges from 2-3dB for fractional power control – consider 3dB by default in LTE Link Budget 106 | Presentation Title | Month 2008

PS 384


2.0 dB 2.6 dB

2.2 dB 2.6 dB

2.0 dB 2.4 dB

2.0 dB 2.3 dB

Uplink Link Budget What Determines the IoT for LTE?

 The average IoT is dependent upon the targeted cell edge data rate (SINR)  The higher the cell edge SINR target, the higher the average IoT

Based on system level simulations:

100.0 dB

 Omni and Directional UE antennas IoT

 SINRs resulting in an IoT > 5-6dB is not considered reasonable

10.0 dB

1.0 dB

0.1 dB

Omni UE Antenna

Realistic Cell Edge SINR Operating Range

-6.0 dB

-4.0 dB

-2.0 dB

0.0 dB

Directional UE Antenna

2.0 dB

4.0 dB

Cell Edge SINR Target



6.0 dB

8.0 dB

Uplink Link Budget Overall MAPL & Cell Range

Overall MAPL for a given service:

MAPLjdB  PMaxTXdBm  TxgaindB  TxlossdB  RxgaindB  RxlossdB  BodylossdB  PenetrationdB  Sensitivity dBm  InterferenceMargindB  ShadowingMargindB  HOGaindB Transmit Power

Max UE transmit Power Maximum Allowable Pathloss

Losses and Margins Gains

Interference margin extra cell interference

Gains - Losses- Margins

Reference Sensitivity

Reference Sensitivity

Interference

cell radius

•= MAPL 108 | Presentation Title | Month 2008


Uplink Link Budget

MAPLdB  Min

Example for Multiple Services

Dense Urban (2.6GHz)

VoIP

MAPL   K jdB

1

 K 2logR cell 

PS 64

PS 128

PS 256

PS 384

PS 512

PS 768

PS 1000

PS 2000

64 kbps

128 kbps

256 kbps

384 kbps

512 kbps

768 kbps

1000 kbps

2000 kbps

1 RB

2 RB

3 RB

5 RB

8 RB

10 RB

16 RB

20 RB

40 RB

MCS 6

MCS 6

MCS 8

MCS 10

MCS 10

MCS 10

MCS 10

MCS 10

MCS 10

Used Bandwidth

180 kHz

360 kHz

540 kHz

900 kHz

1440 kHz

1800 kHz

2880 kHz

3600 kHz

7200 kHz

Target C/I

-3.7 dB

-3.6 dB

-3.0 dB

-2.4 dB

-2.9 dB

-3.1 dB

-3.4 dB

-2.9 dB

-3.3 dB

eNode-B Noise Figure

2.5 dB

2.5 dB

2.5 dB

2.5 dB

2.5 dB

2.5 dB

2.5 dB

2.5 dB

2.5 dB

Required Data Rate 12.2 kbps No. Resource Blocks Required MCS

eNode-B Sensitivity -122.7 dBm -119.6 dBm -117.2 dBm -114.4 dBm -112.9 dBm -112.1 dBm -110.3 dBm -108.8 dBm -106.2 dBm Antenna Gain

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

18.0 dBi

0.5 dB

0.5 dB

0.5 dB

0.5 dB

0.5 dB

0.5 dB

0.5 dB

0.5 dB

0.5 dB

Body Losses

3 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

Additional UL Losses

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

Cell area coverage probability

95%

95%

95%

95%

95%

95%

95%

95%

95%

Overall standard deviation

8.0 dB

8.0 dB

8.0 dB

8.0 dB

8.0 dB

8.0 dB

8.0 dB

8.0 dB

8.0 dB

Shadowing Margin

8.6 dB

8.6 dB

8.6 dB

8.6 dB

8.6 dB

8.6 dB

8.6 dB

8.6 dB

8.6 dB

Handoff Gain

3.6 dB

3.6 dB

3.6 dB

3.0 dB

2.4 dB

2.0 dB

1.5 dB

1.1 dB

0.5 dB

Fast Fading Margin

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

0 dB

Penetration Margin

21 dB

21 dB

21 dB

21 dB

21 dB

21 dB

21 dB

21 dB

21 dB

Fixed IoT

3.0 dB

3.0 dB

3.0 dB

3.0 dB

3.0 dB

3.0 dB

3.0 dB

3.0 dB

3.0 dB

0 dBi

0 dBi

0 dBi

0 dBi

0 dBi

0 dBi

0 dBi

0 dBi

0 dBi

23 dBm

23 dBm

23 dBm

23 dBm

23 dBm

23 dBm

23 dBm

23 dBm

23 dBm

131.1 dB

128.7 dB

125.3 dB

123.1 dB

122.0 dB

119.7 dB

117.8 dB

114.5 dB

0.53 km

0.46 km

0.37 km

0.32 km

0.30 km

0.25 km

0.23 km

0.18 km

Cable & Connector Losses

UE Antenna Gain UE Max Transmit Power

MAPL 131.2 dB UL Cell Range


0.53 km


Uplink Link Budget Fractional Power Control – Handling in LKB (4/4)

Respecting the SINR slope (dictated by the fractional power control parameters) means for services requiring very high SINR values that:  Substantial reductions in allowable UE transmit power are required  The corresponding impact on the link budget is substantial



Uplink Link Budget Propagation Models

For 700, 850 or 900 MHz - Okumura-Hata:  K1 = 69.55 + 26.16 x log10(FMHz) - 13.82 x log10(Hb) - a(Hm) + Kc  a(Hm) = (1.1 x log10(FMHz) - 0.7) x Hm - (1.56 x log10(FMHz) - 0.8) medium-sized city

 K2 = 44.9 -6.55*log10(Hb) For AWS, 1.9GHz or 2.1GHz - COST-231 Hata:  K1 = 46.3 + 33.9 x log10(FMHz) - 13.82 x log10(Hb) - a(Hm) + Kc  K2 = 44.9 - 6.55 x log10(Hb) For 2.6GHz - modified COST-231 Hata: as COST-231 Hata is limited to 1.5GHz to 2GHz  Based on measurements at higher frequencies (2.5GHz & 3.5GHz):  K1 = 46.3 + 33.9 x log10(2000) + 20 x log10(FMHz/2000) - 13.82 x log10(Hb) a(Hm) + Kc  K2 = 44.9 - 6.55 x log10(Hb) 111 | Presentation Title | Month 2008


Uplink Link Budget Impact of TMA (1/3)

Tower Mounted Amplifier (TMA) also called Mast Head Amplifier (MHA)

Vertical Polarisation

Jumper Cable

Impact on link budget

Dual TMA

 Slightly Reduce the global Noise Figure  Compensate the cable losses  0.4dB DL insertion losses

Antenna

Duplexer

Duplexer

Feeder LNA Duplexer

LNA Duplexer

Jumper Cable

TX / RX

Usage recommended for UL coverage-limited scenarios

TXdiv / RXdiv

eNode-B



Tower Mounted Amplifier Impact of TMA (2/3)

Friis formula to compute the overall noise figure of the receiver chain with TMA:

noverall  nTMA  nelement

nfeeder  1 nNode B  1  g TMA g TMA  g feeder

NFelement  10 10

With

g element

Gelement  10 10

and

Where NFfeeder =-Gfeeder =Feeder Losses Typical TMA characteristics:

NFTMA =2 dB

GTMA =12 dB

DL Insertion losses = 0.4dB


Typical gain on uplink link budget (Macro site):  2.9dB gain for sites with 3dB cable losses  3.7 dB gain for sites with 4dB cable losses

Typical gain on uplink link budget (RRH site):  0.3dB gain for sites with 0.6dB cable losses

 Note: TMA should not be considered for RRH sites


Tower Mounted Amplifier Impact of TMA (3/3)


PS 128 (no TMA)

PS 128 (TMA)

128 kbps 3 RB MCS 8 540 kHz -3.0 dB 2.5 dB -117.2 dBm 18.0 dBi 3.0 dB 0 dB 0 dB 95% 8.0 dB 8.6 dB 3.6 dB 0 dB 21 dB 3.0 dB 0 dBi 23.0 dBm 126.2 dB 0.39 km

128 kbps 3 RB MCS 8 540 kHz -3.0 dB 2.4 dB -117.3 dBm 18.0 dBi 0.2 dB 0 dB 0 dB 95% 8.0 dB 8.6 dB 3.6 dB 0 dB 21 dB 3.0 dB 0 dBi 23.0 dBm 129.1 dB 0.47 km

Reduced Noise figure (based on Friis formula) No cable losses but 0.2dB jumper losses


Around 2.9dB gain on MAPL for sites with 3dB cable losses

Common & Control Channel Considerations Overview

There are two main common and control channel considerations that should be assessed for an LTE network design to ensure that they will not limit the coverage. These include:  INTERNAL NOTE – Attach Procedure  ACK/NACK Transmission  Either punctured onto the Physical Uplink Shared Channel (PUSCH)

 Or over the Physical Uplink Control Channel (PUCCH)



INTERNAL NOTE – Common & Control Channel Considerations Attach Procedure

This is the procedure that the UE must go through to Attach to an LTE network UE

eNB

PGW

SGW

MME

RACH Preamble (1) Grant and TA (2)

Limiting Message

RRC Connection Request (3) RRC Connection Setup (4) RRC Connection Setup Complete (5) Attach request (6)

Authentication (optional)/ security (7-8) Create Default Bearer Request (9)

No MME Relocation



CDB Request (10)


From a link budget perspective the limiting message from messages 1, 2, 3, 4, 5, 15 and 16 (that involve the air interface) must be considered to assess any link budget constraints UE

eNB

RRC Connection reconfiguration (14)

SGW

MME

Attach accepted (13)

Create Default Bearer Response (12)

RRC Connection reconfiguration complete (15) Attach complete (16) 1st UL bearer packet

Update Bearer Request (20) Update Bearer Response (21) 1st DL bearer packet

No MME Relocation



CDB Response (11)

PGW


Message 3 (RRC Connection Request)  1 resource block with QPSK rate 1/3 providing an average effective data rate of 20.8 kbps (after 5 HARQ transmissions)  SINR requirement = 0.7dB (including margins) UL link budget  Dense Urban  2.6GHz band

Attach LKB Can be Limiting Depending on Cell Edge Rate Target



Common & Control Channel Considerations ACK/NACK Transmission

DL transmission requires a steady stream of ACK transmissions over the UL to acknowledge the DL packets  Correct ACK reception is critical for optimizing the DL efficiency ALU punctures ACK over the PUSCH initially and over the PUCCH in the longer term ACK/NACK Transmission:  1 RB, QPSK, SINR -3.4dB (PUSCH) & -4.2dB (PUCCH) UL LKB for Urban, 2.6GHz band ACK Is Never Foreseen to Limit UL Coverage 119 | Presentation Title | Month 2008


LTE Link Budgets Downlink Link Budget Considerations



Downlink Link Budget Rationale Behind Downlink LKB Formulation (1/3)

1. DL Cell range defined by UL cell edge service link budget 2. DL throughputs computed for coverage probabilities associated with each corresponding UL service 3. Geometry distribution used for determining the cell edge throughput



Downlink Link Budget Rationale Behind Downlink LKB Formulation (2/3) UL Rates 128kbps (3RB) - guaranteed at cell edge 256kbps (5RB) 512kbps (10RB)

RangeUL_Guar_Serv

8623kbps (50RB) 3921kbps (50RB) 1323kbps (50RB) DL Rates

The above example illustrates the detailed DL Link Budget on the subsequent slides … Urban morphology, indoor 0dBi omni UE configuration, cell range fixed for UL 128kbps, 100% adjacent cell DL RB Loading, No TMA Note: The diagram is not to scale and doesn’t include all rates 122 | Presentation Title | Month 2008


Downlink Link Budget Rationale Behind Downlink LKB Formulation (3/3)

 Uniform power per RB is assumed on the DL  DL performances extracted from link level simulations  The optimal MCS is selected for given number of RB to maximize throughput while ensuring a 20% initial BLER

 Only TxDiv is assumed for referenced DL link level simulations  As the DL link budget is focusing on cell edge performances it is considered that the rank and geometry are insufficient to justify Spatial Multiplexing (SM)  Where a relatively low rate is guaranteed on the UL at cell edge, e.g. 512kbps) the relative UL cell ranges for the high UL rates will be very small and thus the corresponding DL SINRs will be relatively high due to the reduced coverage reliability – in such cases there is some justification for consideration SM performances (not yet incorporated here)



Downlink Budget Example: 10MHz BW Dense Urban (2.6GHz) No. Resource Blocks Used Bandwidth UE Noise Figure eNode-B Antenna Gain Cable & Connector Losses Body Loss Penetration Margin Limiting UL Cell Range # DL Tx Paths Total DL eNode-B Tx Power / Path % DL Power for PDSCH Max eNode-B Tx Power / Service UE Antenna Gain Adjacent Cell Loading UL Service Cell Range DL Path Loss @ UL Cell Edge Total DL Losses @ UL Cell Edge DL Cell Area Coverage Probability Geometry at UL Service Cell Range Desired Signal Adjacent Cell Signal Noise Cell Edge SINR Optimal MCS Data Rate at UL Service Cell Edge 124 | Presentation Title | Month 2008

PS 128

PS 256

50 RB 9000 kHz 7 dB 18 dBi 0.5 dB 0 dB 21 dB 0.46 km 2 paths 30 W 80% 46.8 dBm 0 dBi 100% 0.46 km 129.1 dB 150.6 dB 95% -4.9 dB -85.8 dBm -80.9 dBm -97.5 dBm -5.0 dB MCS 2 1323 kbps



Equivalent UL Service

Cell Range for Limiting UL Service (128kbps)

Cell Range for Equivalent UL Service (256kbps)

Coverage Probability for DL service  95% x (0.36)2 / (0.46)2


PS 128

PS 256




% of total DL power dedicated to PDSCH

Geometry at the corresponding UL service range

The cell edge SINR

Downlink Budget DL Power Settings

Depending on the OAM power offset settings for the Resource Elements (RE) of different channel types we can compute the Average PDSCH Power / OFDM Symbol  Example below for 10MHz, 2 x 40W PA Power  Average % power / symbol allocated to PDSCH RE’s  32.1 / 40 = 80.2%



Downlink Budget Geometry & SINR (1/2) Geometry distributions from system simulations

Geometry 

 A range of UE configurations, both

Rx Power Serving Site

 Rx Power

Adjacent Site

All

omni and, directional UEs (fixed wireless)  Examples in LKB are for coverage

Geometry Distributions (Different UE Configs)

reliabilities of 95% and 61%

100% 90%

 Yield Geometries of -3.9 & 4.7dB

80%

respectively

70%

CDF

61% Coverage Reliability 60%

An additional 1dB is subtracted from these geometry values to align with field expectations

50%

Outdoor - 2 dBi - Omni Outdoor - 4 dBi - Omni Outdoor - 4 dBi - Direc. Outdoor - 6 dBi - Direc. Outdoor - 8 dBi - Direc. Outdoor - 10 dBi - Direc. Indoor - 0 dBi - Omni Indoor - 2 dBi - Omni Indoor - 4 dBi - Omni

40% 30% 20%

95% Coverage Reliability

10% 0% -5.0 dB

-1.0 dB

Geometry -3.9dB 127 | Presentation Title | Month 2008

3.0 dB

7.0 dB

11.0 dB

Geometry Geometry 4.7dB


15.0 dB

19.0 dB

23.0 dB

Downlink Budget Geometry & SINR (2/2)

PDSCH SINR for a defined cell range and coverage reliability:  PDSCHSINR = PDSCHRx / [ PDSCHRx – Geometry + Thermal Noise] Where:  PDSCHRx = PowerPDSCH – Total DL Losses  PowerPDSCH = PowerMax PA x Power FractionPDSCH x RBService / RBMax – Power FractionPDSCH is the average fraction of the total power allocated to PDSCH Resource Elements (REs) per symbol across all RB’s

 Thermal Noise = 10 x log10( F x Nth x NRB x WRB ) – – – –

F: eNode-B Noise figure in dB Nth: Thermal noise density, 10log(Nth) =-174 dBm/Hz NRB: Number of resource blocks (RB) required to reach a given data rate WRB: Bandwidth of one Resource Block




PS 128

PS 256



Max # RB for the bandwidth is assumed by default

The optimal MCS for the #RB and SINR Corresponding L1 Throughput for #RB, MCS and SINR


Downlink Link Budget SINR Performances - Overview

Like the UL the DL SINR Performances depends on:  eNode-B equipment performance

 Radio conditions (multipath fading profile, mobile speed)  Receive diversity (2-way by default or optional 4-way)  Targeted data rate and quality of service

 The Modulation and Coding Scheme (MCS)  Max allowed number of HARQ transmissions  HARQ Operating Point – 20% BLER for 1st HARQ Transmission considered by default Derived from link level simulations Note: Currently the Link Level Simulations referenced in the DL LKB are for EVehA3km/h, 2x2 TxDiv 130 | Presentation Title | Month 2008


Downlink Link Budget SINR - Selection of Optimal SINR Figures

Based on a set of link level simulation results:  Full range of MCS values

 Full range of # RB’s

LTE DL 2x2 MIMO. EVA-3km/hr 60000

50000

Example for Downlink 50RB, 10MHz Bandwidth (2x2 MIMO)

Throughput (kbps)

40000

30000

MCS = 0

MCS = 1

MCS = 2

MCS = 3

MCS = 4

MCS = 5

MCS = 6

MCS = 7

MCS = 8

MCS = 9

MCS = 10

MCS = 11

MCS = 12

MCS = 13

MCS = 14

MCS = 15

MCS = 16

MCS = 17

MCS = 18

MCS = 19

MCS = 20

MCS = 21

MCS = 22

MCS = 23

MCS = 24

MCS = 25

MCS = 26

MCS = 27

MCS = 28

T'put (kbps)

20000

10000

0 -10

-5

0

5

10

15

20

SNR (dB)



25

30

35

40

45

50

Downlink Link Budget Downlink Performance Analysis (1/3)

Downlink Link Level Results for:  25 RB, MCS 28, TxDiv and 5MHz Bandwidth 16000 kbps

120.0%

14000 kbps

12Mbps Throughput

8000 kbps 6000 kbps 4000 kbps

Throughput

60.0%

40.0%

20% BLER

BLER_0

2000 kbps 0 kbps 12.00 dB

14.00 dB

16.00 dB

18.00 dB

20.00 dB

SINR 132 | Presentation Title | Month 2008


22.00 dB

24.00 dB

20.0%

0.0% 26.00 dB

BLER

80.0%

10000 kbps

19.4dB SINR

Throughput

12000 kbps

100.0%


Downlink Link Level Results for:  25 RB, 1-28 MCS, TxDiv and 5MHz Bandwidth

 -5dB cell edge SINR 14000 kbps

25 dB

Throughput 20 dB

SINR

10000 kbps

15 dB

8000 kbps

10 dB

6000 kbps

5 dB

4000 kbps

0 dB

-5dB Cell Edge SINR Target

2000 kbps

-5 dB

660 kbps T’put 0 kbps

MCS 1

0

-10 dB 5

10

15

MCS Index 133 | Presentation Title | Month 2008


20

25

SINR

Throughput

12000 kbps


Downlink Link Level Results for:  1 to 25 RB, All MCS, TxDiv and 5MHz Bandwidth

 -5dB cell edge SINR MCS 6

Throughput

MCS 5 100 kbps

MCS 4 MCS 3

10 kbps

MCS 2 Throughput

MCS 1

Throughput / RB

1 kbps 2 RB

MCS

MCS 0 7 RB

12 RB

17 RB

# Resource Blocks 134 | Presentation Title | Month 2008


22 RB

Modulation & Coding Scheme

1000 kbps

Downlink Budget Example: 10MHz BW (Multiple Services) Dense Urban (2.6GHz) No. Resource Blocks Used Bandwidth UE Noise Figure eNode-B Antenna Gain Cable & Connector Losses Body Loss Penetration Margin Limiting UL Cell Range # DL Tx Paths Total DL eNode-B Tx Power / Path % DL Power for PDSCH Max eNode-B Tx Power / Service UE Antenna Gain Adjacent Cell Loading UL Service Cell Range DL Path Loss @ UL Cell Edge Total DL Losses @ UL Cell Edge DL Cell Area Coverage Probability Geometry at UL Service Cell Range Desired Signal Adjacent Cell Signal Noise Cell Edge SINR Optimal MCS Data Rate at UL Service Cell Edge 135 | Presentation Title | Month 2008

PS 128

PS 256

PS 512



50 RB 9000 kHz 7 dB 18 dBi 0.5 dB 0 dB 21 dB 0.46 km 2 paths 30 W 80% 46.8 dBm 0 dBi 100% 0.30 km 122.4 dB 143.9 dB 40% 3.3 dB -79.1 dBm -82.4 dBm -97.5 dBm 3.2 dB MCS 10 8623 kbps


Downlink Link Budget Summary

The downlink link budgets presented here are indicative of what rates are achievable within the corresponding UL service coverage areas  LTE coverage is not considered to be limited by the DL  for typical eNode-B output powers and deployment scenarios with a 23dBm UE output power, link budgets should remain uplink limited It is important to understand that:

 DL cell edge performances are strongly dependent upon scheduler parameters (e.g. tuning of the fairness of the proportional fair scheduler algorithm) or the available bandwidth (e.g. 10MHz vs 5MHz)  DL performances in the link budget are based only on long term average PDSCH SINR values and do not account for dynamic channel variations that can be addressed with frequency selective scheduling functionalities Better estimates of DL performances can be achieved by means of:  System level simulations and/or Radio Network Planning (RNP) analysis 136 | Presentation Title | Month 2008


Downlink Link Budget Required DL Output Power ?

A series of system simulation studies were performed to assess the required Power Amplifier (PA) sizing for 3 different important cases  700 MHz (10 MHz), 2.1 GHz (10 MHz), 2.1 GHz/AWS (5 MHz) and 2.6 GHz (20 MHz)  All scenarios considered 2x2 MIMO on the DL and 2RxDiv on the UL In principle, all studies concluded the following:  Spectrum efficiency for “reasonable” cell sizes is relatively invariant to reasonable choices for PA sizes  Edge rates become much more sensitive to the choice of power at large cell radiuses



Downlink Link Budget Downlink PA Sizing for LTE – Conclusions

Recommendations from study (independent of frequency)


Carrier Bandwidths

PA Power

1.4 MHz

2 x 10 W

3.0 MHz

2 x 10 W

5.0 MHz

2 x 20 W

10.0 MHz

2 x 30 W

15.0 MHz

2 x 40 W

20.0 MHz

2 x 40 W


RF Design



LTE eNode-B Dimensioning Key Issues to be considered

Coverage

Cell edge coverage expectations + depth of coverage Target operating frequency band + propagation assumptions

Overlay versus Greenfield deployment  Antenna system sharing requirements (impact on coverage + optimization constraints) Radio features, e.g. TMA, RRH, ICIC

Capacity

Subscriber usage profile Subscriber forecast Spectrum constraints Peak throughput requirements Radio features, e.g. ICIC 140 | Presentation Title | Month 2008


Rollout Phase Site Field Positioning Principles

 Based on Site Count (from RF dimensioning process)  Sites positioned to satisfy – RS coverage target (from LB for a target area reliability) – Capacity requirement

 Placed either manually or utilizing Automatic Cell Planning (ACP) tools

 Site Sharing Approach:  The first and quickest approach without RNP is to overlay existing sites with LTE – A 1:1 mapping is most appropriate where the overlaid network is at a frequency band close to LTE band

 Site overlay optimized with the aid of RNP predictions with an accurate propagation model – Sites can be added or deleted where there is limited or excess coverage, respectively – Analysis performed at the same time as antenna azimuth optimization (see next slide)



Rollout Phase RF Optimization Criteria

 Azimuth optimization and tilt optimization are the main rules to optimize the network in order to have the best radio environment before implementing any features.

 The aim are  Optimize coverage in order to reach RSRP targets

 To reduce the number of servers covering the same area in order to avoid excessive overlapping. – This minimize interference without impacting coverage, improve SINR so network performances like – Throughput – Capacity – Frequency re-use efficiency



Rollout Phase RSRP target

 RS-RSSI: total power transmitted dedicated for Reference signal during one OFDM symbol duration  Currently in Atoll it is more RS-RSSI is calculated, and the total power dedicated to RS is 1/6 of Max power. This approach is not 100% of the time in line wit power settings on the field  LA0.x for a 30W PA power energy per RE for RS is 14.9 dBm. Considering 10MHz bandwidth 100 RE are used to calculate RS-RSSI, so total power dedicated to RS over one OFDM symbol is 34.9dBm, but Atoll calculates 30W/6, so 37dBm, so to do the right calculation for this configuration max power set in Atoll should be 43dBm instead of 45dBm.



Rollout Phase RSRP target  LA1.0 for RRH 30W PA power energy per RE for RS is 16.2 dBm. Considering 10MHz bandwidth 100 RE are used to calculate RS-RSSI, so total power dedicated to RS over one OFDM symbol is 36.2dBm, but Atoll calculates 30W/6, so 37dBm, so to do the right calculation for this configuration max power set in Atoll should be 44dBm instead of 45dBm.  LA1.0 for TRDU 40W PA power energy per RE for RS is 18.2 dBm. Considering 10MHz bandwidth 100 RE are used to calculate RS-RSSI, so total power dedicated to RS over one OFDM symbol is 38.2dBm, Atoll calculates 40W/6, so 38dBm, so it is ok

 3GPP RSRP definition:  Reference signal received power (RSRP), is determined for a considered cell as the linear average over the power contributions (in [W]) of the resource elements that carry cell-specific reference signals within the considered measurement frequency bandwidth.



Rollout Phase RF Optimization Criteria Outdoor RSRP target depending on environment and frequencies for UL PS 128 service and UL PS 256, considering 45dBm PA power and 14.9 dBm Reference signal Tx power per RE. RSRP value does not depends on the number of transmit

 DL RS EIRP per RE and per transmit:  30.9dBm @ 2600MHz/2100MHz/AWS/1900MHz/1800MHz with 18dBi antenna gain & 2dB cable losses  30.9dBm @ 900MHz/850MHz with 17dBi antenna gain & 1dB cable losses  28.9dBm @700MHz with 15dBi antenna gain & 1 dB cable losses



Rollout Phase RF Optimization Criteria

 Currently the calculation done in 9155 is the sum of all Reference signal resource elements power transmitted in a same OFDM time period over all the bandwidth. This approach is not in line with 3GPP as 3GPP specify the linear average of reference signal resource elements.  To compensate this error the following work around must be followed and based on the same analysis done for RS-RSSI calculation  LA0.x for RRH 30W PA power energy per RE for RS is 14.9 dBm. – For 5MHz bandwidth set in Cell table Max power column: eNode-B PA power -19dB – For 10MHz bandwidth set in Cell table Max power column: eNode-B PA power -22dB – For 20MHz bandwidth set in Cell table Max power column: eNode-B PA power -25dB



Rollout Phase RF Optimization Criteria  LA1.0 for RRH 30W PA power energy per RE for RS is 16.2 dBm. – For 5MHz bandwidth set in Cell table Max power column: eNode-B PA power -18dB – For 10MHz bandwidth set in Cell table Max power column: eNode-B PA power -21dB – For 20MHz bandwidth set in Cell table Max power column: eNode-B PA power -24dB

 LA1.0 for TRDU 40W PA power energy per RE for RS is 18.2 dBm. – For 5MHz bandwidth set in Cell table Max power column: eNode-B PA power -17dB – For 10MHz bandwidth set in Cell table Max power column: eNode-B PA power -20dB – For 20MHz bandwidth set in Cell table Max power column: eNode-B PA power -23dB



Rollout Phase RF Optimization Criteria  The method proposed is to:  Set indoor penetration losses in 9155 clutter table  Use the UL Link Budget Available Path loss with 0dB penetration losses set in the LB for the dimensioning service selected,

 Design RSRP = RS per RE EIRP+ ANT_GAIN – Available Uplink Pathloss – indoor losses  where: – RS per RE EIRP = Reference signal EIRP per resource element , it is automatically calculated by 9155 when the work around specified above is followed – ANT_GAIN = Node-B antenna gain – Available Uplink Pathloss: UL available pathloss calculated with the link budget when penetration loss is set to 0dB

 The RSRP target values specified in slide , have been defined with this approach.  If the user apply this approach, the following recommendation must be respected  Select “indoor loss” icon in 9155 coverage study Do not select “shadowing taken into account “ icon as it is already done in RSRP target calculated below



RF optimization criteria  Overlapping optimization  The following rules are not technology specifics, and their efficiency have already been measured on GSM, W-CDMA networks.

 Pollution and interference analysis – Within 4dB of the best server – number of servers should ≤ 4 – % area with 4 servers should be < 2%. – % of area with 2 servers should be < 30%. – Within 10dB of the best server – number of servers should ≤ 7 – % of area with 7 servers should be < 2%. – High signal level overlap analysis: – Increase the design threshold for the covered area by 10dB – % of 3 servers in the design area should not exceed 10%.. – Example: if the RS design threshold is -85dBm, a number of server’s analysis is done with a threshold equal to -75dBm.



RF optimization criteria  SINR target  This target can be used with 9155 RNP tool, but it is not 100% sure that it can be measured on the field with high accuracy as it is not 3GPP measurement criteria.

 In 9155 SINR can be calculated based on reference signal, or PDSCH, and for loaded cases it provides the same results as power per RE RS= power per RE PDSCH  The SINR target value depends on the traffic load: – 95% of the design area should have SINR ≥ -5dB, with 100% DL load – 95% of the design area should have a SINR ≥-2dB with 50% DL load

 SINR does not depends on number of transmits



RF optimization criteria  RSRQ target  RSRQ= N*RSRP/RSSI where RSSI is all the power received in the N resource blocks used bandwidth during the same time period where RSRP is measured.  RSRQ depends on the number of transit, as RSSI value depends on it, and not RSRP  RSRQ target value depends on the traffic load:  1 transmit : – 95% of the design area should have RSRQ ≥ -17dB, with 100% DL load – 95% of the design area should have RSRQ ≥ -14dB, with 50% DL load

 2 transmits : – 95% of the design area should have RSRQ ≥ -20dB, with 100% DL load – 95% of the design area should have RSRQ ≥ -17dB, with 50% DL load

 4 transmits : – 95% of the design area should have RSRQ ≥ -23dB, with 100% DL load – 95% of the design area should have RSRQ ≥ -20dB, with 50% DL load



RF optimization criteria  These targets are been obtained on several well known environments ; where a very good optimization has been done in W-CDMA due to critical inter-site distance : 400m. Same RNP environment has been re-used for LTE predictions without changing anything to evaluate the best SINR & RSRQ reachable in different full traffic load condition.  The RNP prediction and RF optimization done for the different trials in US and Europe confirm that these targets can be reach and are a good way to optimize throughput and reduce interferences.  Overlapping criteria, RSRQ target and SINR target defined above are in line to provide the same RF design. They allow managing interferences in order to obtain a RF network design able to support the best throughput .  10Mbps in cell center for mono-user when all surrounded cells have 100% load  1.5Mbps at cell edge in mono-user for 10MHz bandwidth when all surrounded cells have 100% load



RF optimization criteria  Neighbors & Cell ID planning criteria  Cell id is required to identify each cell, a cell id is the combination of one of the 3 sequences supported by P-SCH and the group Id supported by S-SCH. – So Realizing a cell id planning = realizing P-SCH planning and S-SCH planning – The strategy recommended is to use the same S-CH per site which induces that each sector uses a different P-SCH sequence

 This distance depends on propagation path loss, the environment and the frequency.  The main criteria are the following one:

 Considering two cells cell A and cell B, on the same frequency carrier using the same cell ID, the distance between those must satisfy the following criterias: – RSRP criteria – At cell A edge (RSRPcellA ≤ -115dBm) : RSRPcellA ≥ : RSRPcellB + 10dB – At cell B edge (RSRPcellB ≤ -115dBm): RSRPcellB ≥ : RSRPcellA + 10dB – RSRQ criteria for 100% load case ( 2 transmits) – At cell A edge (RSRQcellA ≤ -20dB) : RSRQcellA ≥ : RSRQcellB + 10dB – At cell B edge (RSRQcellB ≤ -20dB): RSRQcellB ≥ : RSRQcellA + 10dB



RF optimization criteria  Distance criteria  Dense urban/ urban – – – –

2km @ 2600MHz considering 600m cell radius 2,4km @ 1800MHz and 2100MHz considering 700m cell radius 5,5km @ 850MHz and 900MHz considering 1,7km cell radius 6Km @ 700MHz considering 1,9km cell radius

 Suburban – – – –

6km @ 2600MHz considering 1,8km cell radius 7km @ 1800MHz and 2100MHz considering 2,2km cell radius 18km @ 850MHz and 900MHz considering 5,5km cell radius 20Km @ 700MHz considering 6km cell radius

– – – –

17km @ 2600MHz considering 6km cell radius 21km @ 1800MHz and 2100MHz considering 7km cell radius 60km @ 850MHz and 900MHz considering 18km cell radius 65Km @ 700MHz considering 20km cell radius

 Rural



www.alcatel-lucent.com www.alcatel-lucent.com



Hard Handover



Hard Handover



Hard Handover Preparation Phase



Hard Handover Execution Phase



Hard Handover Completion phase



Hard Handover Execution time



210185358 LTE Training RF Design Ed4

Recommend Documents