LTE Link Budget (Ejemplo)

ITU/BDT Arab Regional Workshop on “4G Wireless Systems” - Tunisia 2010

ITU/BDT Arab Regional Workshop onITU/BDT Arab Regional Workshop on“4G Wireless Systems”

LTE Technology

Session 5 : LTE Technology PerofrmanceEvaluationEvaluation

Speakers M. Lazhar BELHOUCHETM Hakim EBDELLIM. Hakim EBDELLI

Date 27 – 29 January 2010

www.cert.tn1 LTE Technology Perofrmance Evaluation


Agenda

– Frequency Bands and Typical deployment areas

– Layer 1 Peak Bit Rates

– Terminal Categories

– Reference Sensitivity

– Link Budgets

– Propagation model

– Uplink link budgetUplink link budget

– Downlink link budgets

– Comparison between GSM/HSPA/LTE

– Latency

– LTE Refarming to GSM Spectrum

i i i i

www.cert.tn2

– Capacity Dimensioning

LTE Technology Perofrmance Evaluation


Introduction

• Performance evaluation : LTE capabilities from the end user’s and from the operator’s point of viewp p

• The operator is interested in the network efficiency: – how many customers can be served,

– how much data can be provided and how many base station sites are required.

Th d li i f d d• The end user application performance depends on :– available bit rate,

– latency and– latency and

– seamless mobility



Frequency Bands



Frequency Bands : Typical deployment areas

• Europe:– Band 7:The 2.6 GHz auctions have been running in a few countries during 2007 and

2008, and continue during 2009/2010.

– Band 8:is currently used mostly by GSM. The band is attractive from a coverage point of view due to the lower propagation losses. The band can be reused for LTE or for HSPA.

– Band 3: is also used by GSM, but in many cases Band 3 is not as heavily used by GSM as Band 8. That makes refarming for LTE simpler.

– Digital dividend

• USA: Bands 4, 12, 13, 14 and 17. Bands 2 and 5 can be used for LTE refarming.

• Japan : Bands 1, 9, 11 and 18.

• LTE deployments globally will use several different frequency bands from the start.




PERFORMANCE EVALUATIONLTE Technology Perofrmance Evaluation



Layer 1 Peak Bit Rates

sSubCarrierofNumberHzbits

msSubFramepersymbolsofNumbersMbratebitPeak *)(*

1]/[ =

• control and reference signal overheads:– PDCCH: takes one symbol out of 14 symbols. That is the minimum possible PDCCH

allocation The resulting control overhead is 7 1% ( = 1/14)allocation. The resulting control overhead is 7.1% ( = 1/14).

– Downlink Reference Signals (RS) depend on the antenna configuration :Single stream transmission uses 2 RS out of 14 in every 3rd sub‐carrier, 2 × 2 MIMO 4 symbols and 4 × 4 MIMO 6 symbols The overhead varies between 4 8% and 14 3%symbols and 4 × 4 MIMO 6 symbols. The overhead varies between 4.8% and 14.3%.

– Other downlink symbols Overhead: synchronization signal, PBCH, PCFICH, and one group of PHICH. The overhead depends on the BW ranging from below 1% at 20 MHz to approximately 9% at 1 4 MHzMHz to approximately 9% at 1.4 MHz.

– Uplink reference signals take 1 symbol out of 7 symbols resulting in an overhead of 14.3% = 1/7.



Layer 1 Peak Bit Rates – Cont.

BW RB/Sub‐carriers

/Modulationand coding

Bits/symbol

MIMO usage 1.4 MHz6/72

3.0 MHz15/180

5.0 MHz25/300

10 MHz50/600

15 MHz75/900

20 MHz100/1200

QPSK 1/2 1 Single stream 0.9 2.2 3.6 7.2 10.8 14.4

16QAM 1/2 2 Single stream 1.7 4.3 7.2 14.4 21.6 28.8


64QAM 3/4 4.5 Single stream 3.9 9.7 16.2 32.4 48.6 64.8


64QAM 3/4 9 2 × 2 MIMO 7.8 19.4 32.4 64.8 97.2 129.6

64QAM 1/1 12 2 × 2 MIMO 10.4 25.9 43.2 86.4 129.6 172.8Q /

64QAM 1/1 24 4 × 4 MIMO 20.7 51.8 86.4 172.8 259.2 345.6

Downlink peak bit rates (Mbps)


p ( p )


Layer 1 Peak Bit Rates – Cont.

BW RB/Sub‐carriers

Modulation Bits/ MIMO usage 1 4 MHz 3 0 MHz 5 0 MHz 10 MHz 15 MHz 20 MHzModulationand coding

Bits/symbol

MIMO usage 1.4 MHz6/72

3.0 MHz15/180

5.0 MHz25/300

10 MHz50/600

15 MHz75/900

20 MHz100/1200

QPSK 1/2 1 Single stream 0.9 2.2 3.6 7.2 10.8 14.4



16QAM 1/1 4 Single stream 3 5 8 6 14 4 28 8 43 2 57 616QAM 1/1 4 Single stream 3.5 8.6 14.4 28.8 43.2 57.6

64QAM 3/4 4.5 Single stream 3.9 9.7 16.2 32.4 48.6 64.8


Uplink peak bit rates (Mbps)



Terminal Categories

Category 1 Category 2 Category 3 Category 4 Category 5

Peak rate downlink (approximately) 10 Mbps 50 Mbps 100 Mbps 150 Mbps 300 Mbps

Peak rate uplink (approximately) 5 Mbps 25 Mbps 50 Mbps 50 Mbps 75 Mbps

Max bits received within TTI 10 296 51 024 102 048 149 776 299 552

Max bits transmitted within TTI 5 160 25 456 51 024 51 024 75 376

RF bandwidth 20 MHz 20 MHz 20 MHz 20 MHz 20 MHz

Modulation downlink 64QAM 64QAM 64QAM 64QAM 64QAMModulation downlink 64QAM 64QAM 64QAM 64QAM 64QAM

Modulation uplink 16QAM 16QAM 16QAM 16QAM 64QAM

Receiver diversity Yes Yes Yes Yes Yes

MIMO downlink Optional 2 × 2 2 × 2 2 × 2 4 × 4



Reference Sensitivity

• The reference sensitivity level is the minimum mean

• kTB :thermal noise level , in units of dBm, in the specified bandwidth (B),

received signal strength applied to antenna ports at

• NF :noise figure for the receiver,

• SINR is the signal to interference plus noise requirement for the chosenwhich there is sufficient SINR

for the specified modulation scheme to meet a minimum

noise requirement for the chosen modulation and coding scheme,

• IM is the implementation marginscheme to meet a minimum throughput requirement of 95% of the maximum

• −3 dB represents the diversity gain.

possible.)(3Re dBmIMSINRNFKTBfSens −+++=



Reference Sensitivity

‐70.00

E‐UTRA reference sensitivity

‐85.00

‐80.00

‐75.00

‐100.00

‐95.00

‐90.00

dBm

‐110.00

‐105.00

1.4 3 5 10 15 20

BandWidth [MHz]BandWidth [MHz]

QPSK 1/8 QPSK 1/5 QPSK 1/4 QPSK 1/3 QPSK 1/2

QPSK 2/3 QPSK 3/4 QPSK 4/5 16QAM 1/2 16QAM 2/3

16QAM 3/4 16QAM 4/5 64QAM 2/3 64QAM 3/4 64QAM 4/5


Q / Q / Q / Q / Q /


Link Budgets

• The link budget calculations estimate the maximum allowed signal attenuation between the mobile and the base station gantenna. The maximum path loss allows the maximum cell range to be estimated with a suitable propagation model.

• The cell range gives the number of base station sites required to cover the target geographical area.



Link Budgets ‐ Propagation model

• A propagation model describes the average signal propagation, and it converts the maximum allowed p p g ,propagation loss to the maximum cell range.

• It depends on:– Environment : urban, rural, dense urban, suburban, open, forest, sea…

– Distance

– Frequency

– atmospheric conditions

– Indoor/outdoor– Indoor/outdoor

• Examples : Free space, Walfish–Ikegami, Okumura–Hata, Longley–Rice, Lee, Young

www.cert.tn14

g y , , g



Link Budgets ‐ Propagation model –Cont.

• Hata Model for Urban Areas:)log(*)]log(*556944[)log(*8213)log(*16265569 dhChfLu −+−−+=

• For small or medium sized city,

)log()]log(55.69.44[)log(82.13)log(16.2655.69 dhChfLu BHB ++=

)log(*56.1*)7.0)log(*1.1(8.0 fhfC MH −−+=

• For large cities,

)log(56.1)7.0)log(1.1(8.0 fhfC MH +

⎧ ≤≤− 200150,1.12))*54.1(log(*29.8 fifhM

⎩⎨⎧

≤−≤≤

=1500200,97.42))*75.11(log(*2.3

200150,1.12))54.1(log(29.8fifh

fifhC

M

MH

p

Parameters Unit Significance Parameters Unit SignificanceLu dB Path loss in Urban Areas f MHz Frequency of TransmissionhB m Height of base station Antenna CH dB Antenna height correction factorhM m Height of mobile station Antenna d km Distance between Base station and MS



Link Budgets ‐ Propagation model –Cont.

• Coverage evaluation will be based on the the Ok H t ti d l ith thOkumura–Hata propagation model with the parameters shown below:

Urban indoor Suburban indoor Rural outdoor Rural outdoor fixedBase station antenna height (m) 30 50 80 80Mobile antenna height (m) 1.5 1.5 1.5 5Mobile antenna gain (dBi) 0 0 0 0Slow fading standard deviation (dB) 8 8 8 8Location probability 95% 95% 95% 95%Correction factor (dB) 0 −5 −15 −15( )Indoor loss (dB) 20 15 0 0Slow fading margin (dB) 8.8 8.8 8.8 8.8



Uplink link budgetsGSM voice HSPA LTE

Data rate (kbps) 12.2 64 64Transmitter – UE

( )

Source :

LTE for UMTS –a Max tx power (dBm) 33 23 23b Tx antenna gain (dBi) 0 0 0c Body loss (dB) 3 0 0d EIRP (dBm) 30 23 23

OFDMA and SC-FDMA Based Radio AccessHarri Holma and Antti Toskala

Receiver – BTS/Node B/eNBe Node B noise figure (dB) – 2 2f Thermal noise (dB) – ‐108.2 ‐118.4g Receiver noise (dBm) – ‐106.2 ‐116.4g ( )h SINR (dB) – ‐17.3 ‐7i Receiver sensitivity ‐114 ‐123.4 ‐123.4j Interference margin (dB) 0 3 1k Cable loss (dB) 0 0 0k Cable loss (dB) 0 0 0l Rx antenna gain (dBi) 18 18 18m Fast fade margin (dB) 0 1.8 0n Soft handover gain (dB) 0 2 0

M i th l 162 161 6 163 4


Maximum path loss 162 161.6 163.4


Downlink link budgetsDownlink GSM voice HSPA LTEData rate (kbps) 12.2 1024 1024Transmitter – Node B

Source :

LTE for UMTS –a Tx power (dBm) 44.5 46 46b Tx antenna gain (dBi) 18 18 18c Cable loss (dB) 2 2 2d EIRP (dBm) 60 5 62 62

OFDMA and SC-FDMA Based Radio AccessHarri Holma and Antti Toskala

d EIRP (dBm) 60.5 62 62Receiver – UE

e UE noise figure (dB) – 7 7f Thermal noise (dB) ‐119.7 ‐108.2 ‐104.5g Receiver noise fl oor (dBm) – ‐101.2 ‐97.5h SINR (dB) – ‐5.2 ‐9i Receiver sensitivity (dBm) ‐104 ‐106.4 ‐106.5j Interference margin (dB) 0 4 4j g ( )k Control channel overhead (%) 0 20 20l Rx antenna gain (dBi) 0 0 0m Body loss (dB) 3 0 0

M i th l 161 5 163 4 163 5


Maximum path loss 161.5 163.4 163.5


Link budgets Comparison GSM/HSPA/LTE

180

Maximum path loss values for GSM voice and HSPA and LTE data

170

175

155

160

165

dB

145

150

155

140

GSM voice HSPA 1Mbps/64kbps LTE 1Mbps/64kbps

Uplink Downlink



Link budgets Comparison GSM/HSPA/LTE• The LTE link budget in downlink has several similarities with HSPA and the

maximum path loss is similar.

h l k h d ff ll f• The uplink part has some differences: smaller interference margin in LTE, no macro diversity gain in LTE and no fast fading margin in LTE.

• The link budgets show that LTE can be deployed using existing GSM and HSPA sites assuming that the same frequency is used for LTE as for GSM and HSPA.

• LTE itself does not provide any major boost in the coverage. That is because the transmission power levels and the RF noise figures are also similar in GSM and HSPA technologies, and the link performance at low data rates is not much different in LTE than in HSPA.

• The link budget was calculated for 64 kbps uplink, which is likely not a high enough data rate for true broadband service. If we want to guarantee higher data rates in LTE, we may need low frequency deployment, additional sites, active antenna solutions or local area solutions.



Cell range Comparison

• The urban cell range varies from 0.6 km to 1.4 km and suburban Cell ranges with Okumura–Hata

propagation modelfrom 1.5 km to 3.4 km.

• Such cell ranges are also typically found in existing GSM and UMTS

100

propagation model

gnetworks.

• The rural case shows clearly higher cell ranges: 26 km for the

10

Km

higher cell ranges: 26 km for the outdoor mobile coverage and even up to 50 km for the rural f d ll

1

fixed installation at 900 MHz. 0.1

900 MHz 1800 MHz 2100 MHz 2600 MHz

Urban indoor Suburban indoor

Rural outdoor Rural outdoor fixed



Difference between 900 MHz and 2600 MHz

Urban RuralRural fixedwireless

• 1 :According to Okumura–Hata.

• 2 :Shared 1.3 m antenna for / d d

Propagation loss (1) +14 dB +14 dB +14 dB

BTS antenna gain ‐3 dB (2) 0 dBc 0 dB (3)

900/2600 giving 15 and 18 dBigain at 900 vs 2600 MHz.

• 3 :2.5 m antenna giving 18 dBi gain at BTS antenna gain 3 dB (2) 0 dBc 0 dB (3)

BTS cable loss (4) +1 dB +3 dB +3 dB

900 MHz.

• 4 :Cable 1/2’. Urban 30 m and rural 100 m.

UE antenna gain ‐5 dB (5) ‐5 dB (5) 0 dB (6)• 5 :Based on 3GPP RAN4

contributions.

• 6 :External fixed antenna assumed. UE sensitivity (7) ‐1 dB ‐1 dB ‐1 dB

Total +6 dB +11 dB +16 dB

• 7 :UE sensitivity can be up to 3 dB worse at 900 MHz but thedifference in practice is less.

www.cert.tn22

difference in practice is less.



Cell range Limitation

• The earth’s curvature limits the maximum cell range to approximately 40 km with an 80 m high base station antenna pp y gassuming that the terminal is at ground level.

• The maximum cell range can be calculated with Equation below

• To achieve 100 km cell range, the required antenna height is 580 m!!!

R = 8650 km hRdR +=+ 222 )((Effective earth radius for radio propagation is4/3 larger than real radius)

RdRh

hRdR

−+=

+=+22

)(



Latency

• User plane latency is relevant for the performance of many applications. pp

• There are several applications that do not require a very high data rate, but they do require very low latency: voice, real time gaming, interactive applications.



Latency : End‐to‐end delay budget

Delay component Delay value

Transmission time uplink + downlink

2 ms

• On average, the packet needs to wait for 0.5 ms for the start of the next framedownlink

Buffering time (0.5 ×transmission time)

2 × 0.5 × 1 ms = 1 ms

Retransmissions 10% 2 × 0.1 × 8 ms = 1.6 ms

frame.

• The retransmissions take 8 ms at best

• the assumed retransmission

Uplink scheduling request 0.5 × 5 ms = 2.5 ms

Uplink scheduling grant 4 ms

UE d l i d 4

probability is 10%.

• The average delay for sending the scheduling request is 2 5 msUE delay estimated 4 ms

eNodeB delay estimated 4 ms

Core network 1 ms

l d l h ll d

scheduling request is 2.5 ms

• Scheduling grant : 4 ms.

• UE processing delay : 4 ms, Total delay with pre‐allocated resources

13.6 ms

Total delay with scheduling 20.1 ms

p g y ,

• eNodeB processing delay :4 ms

• core network delay :1 ms.



End‐to‐end round trip time including scheduling latency

25

LTE round trip time

20

Core

15eNB

UE

Uplink scheduling grant

Uplink scheduling request

5

10Uplink scheduling request

Retransmissions

Buffering time

Transmission time

0

Delay value



LTE Refarming to GSM Spectrum

• LTE could be deployed in the existing GSM spectrum like 900 MHz or 1800 MHzlike 900 MHz or 1800 MHz.

• The flexible LTE bandwidth makes refarming easier th ith WCDMA b LTE t t ith 1 4than with WCDMA because LTE can start with 1.4 MHz or 3.0 MHz bandwidths and then grow later h th GSM t ffi h d dwhen the GSM traffic has decreased.



LTE refarming to GSM spectrum ‐ Example



Dimensioning Capacity ‐ Traffic volume based approach

• The busy hour is assumed to carry 15% of the daily traffic

h b h l d• the busy hour average loading is 50%.

• The calculation shows that the total site throughput per month is 4600 GB.

• To offer 5 GB data for every subscriber per month, the number of subscribers per site will be 920.



Dimensioning Capacity ‐ Data rate based approach

• Target : 1 Mbps per subscriber.

• Since only some of the subscribers are downloading data simultaneously, we can apply an overbooking factor, for example g , p20.

• It means that the average busy hour data rate is 50 kbps perhour data rate is 50 kbps per subscriber.

• The number of subscribers per site using this approach is 1050.



Thanks for your AttentionThanks for your Attention




ANNEXESLTE Technology Perofrmance Evaluation



UL link budget parameters for LTE – Parameters‐1

R f LTE f UMTS OFDMA d SC FDMA B d R di A


Ref. LTE for UMTS – OFDMA and SC-FDMA Based Radio Access Harri Holma and Antti Toskala


DL link budget parameters for LTE – Parameters‐1





DL link budget parameters for LTE – Parameters‐2




LTE Link Budget (Ejemplo)

Documents

Transcript of LTE Link Budget (Ejemplo)