LTE Traffic Model

LTE Radio Access, Rel. RL40, Operating Documentation, Issue 03

LTE Traffic Model DN0972027 Issue 02 Approval Date 2012-10-16 Confidential

Nokia Siemens Networks is continually striving to reduce the adverse environmental effects of its products and services. We w ould like to encourage you as our customer s and users to join us in working towards a cleaner, safer environment. Please recycle product packaging and follow the recommendations for power use and proper disposal of our products and their components. If you should have questions regarding our Environment al Policy or any of the environm ental services we offer, please contact us at Nokia Siemens Networks for any additional information.

LTE Traffic Model

The information in this document i s subject to change without notice and descri bes only the product defined in the introduction of this documentation. This documentation is in tended for the use of Nokia Siemens Networks c ustomers only for the purposes of the agreement under which the document is submitted, and no part of it may be used, reproduced, modified or transmitted in any form or means without the prior written permission of Nokia Siemens Networks. The documentation has been prepared to be used by pr ofessional and properly trained personnel, and the customer assumes full responsibility when using it. Nokia Siemens Networks welcomes customer comments as part of the pr ocess of continuous development and improvement of the documentation. The information or statements given in this documentation concerning the suitability, capacity, or performance of the mentioned hardware or so ftware products are given "as is" and all liability arising in connection with such hardware or software products shall be defined conclusively and finally in a separate agreement between Nokia Siemens Networks and the customer. However , Nokia Siemens Networks has made all reasonable efforts to ensure that the instructions contained in the document are adequate and free of m aterial errors and omissions. N okia Siemens Networks will, if deemed necessary by Nokia Siemens Networks, explain issues whi ch may not be covered by the document. Nokia Siemens Networks will correct errors in this documentation as soon as possible. IN NO EVENT WILL Nokia Siemens Networks BE LIABLE FOR ERRORS IN THIS DOCUMENTATION OR FOR ANY DAMAGES, INCLUDING B UT NOT LIMITED TO SPECIAL, DIRECT, INDIRECT, INCIDENTAL OR CONSEQUENTIAL OR ANY LOSSES, SUCH AS BUT NOT LIMITED TO LOSS OF PROFIT, REVENUE, BUSINESS INTERRUPTION, BUSINESS OPPORTUNITY OR DATA,THAT MAY ARISE FROM THE USE OF THIS DOCUM ENT OR THE INFORMATION IN IT. This documentation and the product it desc ribes are considered protected by copyrights and other intellectual property rights according to the applicable laws. The wave logo is a trademark of Nokia Siemens Networks Oy. Nokia is a registered trademark of Nokia Corporation. Siemens is a registered trademark of Siemens AG. Other product names mentioned in this document may be trademarks of their respective owners, and they are mentioned for identification purposes only. Copyright © Nokia Siemens Networks 2012. All rights reserved

f

Important Notice on Product Safety This product may present safety risks due to laser, electricity, heat, and other sources of danger. Only trained and qualified personnel may i nstall, operate, maintain or otherwise handle this product and only after having carefully read the safety information applicable to this product. The safety information is pr ovided in the Safety Information section in the “ Legal, Safety and Environmental Information” part of this document or documentation set.

The same text in German:

f

Wichtiger Hinweis zur Produktsicherheit Von diesem Produkt können Gefahren durch Laser, Elektrizität, Hitzeentwicklung oder andere Gefahrenquellen ausgehen. Installation, Betrieb, Wartung und sonstige Handhabung des Pr oduktes darf nur durch geschultes und qualifiziertes Personal unter Be achtung der anwendbaren Sicherheitsanforderungen erfolgen. Die Sicherheitsanforderungen finden Sie unter „Sicherheitshinweise“ im Teil „Legal, Safety and Environmental Information“ dieses Dokuments oder dieses Dokumentationssatzes.

2

Id:0900d80580973cde Confidential

DN0972027

LTE Traffic Model

Table of contents This document has 17 pages. Summary of changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

DN0972027

1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 2.1

Traffic dimensionig principles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Overbooking factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3

Traffic modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

4

Default LTE traffic model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Id:0900d80580973cde Confidential

3

LTE Traffic Model

List of figures

4

Figure 1

Overall dimensioning process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Figure 2 Figure 3 Figure 4

Offered vs served traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Traffic daily profile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Overall dimensioning workflow (E-UTRAN + EPC). . . . . . . . . . . . . . . . . 11

Figure 5 Figure 6

Subscription rate vs average throughput . . . . . . . . . . . . . . . . . . . . . . . . 11 Steps in traffic modelling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Id:0900d80580973cde Confidential

DN0972027

LTE Traffic Model

List of tables

DN0972027

Table 1

Exemplary TM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Table 2 Table 3 Table 4

Service characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Traffic mix’s for different subscriber type . . . . . . . . . . . . . . . . . . . . . . . 14 Calculated values for BHCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Table 5

Traffic model for LTE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Id:0900d80580973cde Confidential

5

Summary of changes

LTE Traffic Model

Summary of changes Changes between document issues are cumulative. Therefore, the latest document issue contains all changes made to previous issues. Changes between issues 03 DRAFT (2012-07-13, RL40) and 03 (2012-10-16, RL40)  •

 All chapters: editorial changes

Issue 03 DRAFT (2012-07-13, RL40)

First issue for RL40.

6

Id:0900d80580973afc Confidential

DN0972027

LTE Traffic Model

Introduction

1 Introduction The purpose of this document is to describe the general LTE Traffic Model (TM) concep t and its role in the network dimensioning process. In addition to descriptions of the general conceptual background, the document also provides proposals for a def ault TM for LTE networks that can be used as basic traffic requirement assumptions during dimensioning exercises that have to be executed without information on expected traffic volumes.

DN0972027

Id:0900d80580973cc9 Confidential

7

Traffic dimensionig principles

LTE Traffic Model

2 Traffic dimensionig principles The Traffic Model described in this document is an essential precondition for both LTE Radio Network and LTE Access Transport Network Dimensioning. The objective of Network Dimensioning is:  •

 •

for Radio dimensioning, to determine the number of cells/sites required  •

to cover a certain geographical area (radio coverage dimensioning)

 •

to serve a particular expected total traffic volume demanded by the network's subscribers (radio capacity dimensioning)

for Access/Transport dimensioning, to determine  •

transport resources required for user data, signaling data & O&M data on the LTE/SAE transport interfaces (transport capacity dimensioning)

 A dimensioning campaign can only be started if a particular set of obligatory input data has been collected beforehand. Especially for radio and transport capacity dimensioning, the Traffic Model is one of those essential inputs. For the radio network, the initial point of a dimensioning campaign is to ensure that the desired coverage is achieved (this takes place without any evaluation of expected subscriber traffic volumes or figures). Once this step is made, the minimum amount of needed sites required to cover the area to be served with sufficiently strong radio signal is known.

Figure 1

Overall dimensioning process

Usually the next step is capacity d imensioning. This is because capacity dimensioning basically compares the capacity (= throughput) of a single cell/site with the expected traffic volumes.

8

Id:0900d80580973cca Confidential

DN0972027

LTE Traffic Model

Traffic dimensionig principles

Figure 2 

Offered vs served traffic

The maximum traffic must be managed in the 'busy hour', i t is of essential importance to estimate the amount of traffic that is to be carried by a system/network in the Busy Hour (BH ). This is only one of possible approache s, assuming the worst case from the point of offered traffic. On the other hand, it is the most appropriate solution because the network must be capable of serving users all the time. Figure 3 Traffic daily profile shows the 'busy hour' denotes that time of day, where the highest traffic volumes are to be expected due to subscribers' behavior.

Figure 3

DN0972027

Traffic daily profile

Id:0900d80580973cca Confidential

9

Traffic dimensionig principles

LTE Traffic Model

In a typical network, during a packet call, network subscribers receive and transmit different amounts of data. Moreover, they move/roam from one cell to another. Though the subscribers' traffic demands and roaming behavior during established connections are highly individual and independent events, a statistical representation of t he subscribers' behavior in the network helps planners to achieve suitable figures on which to base a dimensioning campaign. In other words, individual subscriber behavior must be reflected by a modeled 'busy hour traffic profile' (= Traffic Model) the aim of which is to come as close as possible to the traffic volumes that the network in question must carry in reality. The aforementioned aspects are described in this Traffic Model document.  A TM can either be very complex and precise or very simple. Either way, the dimensioning expert should usually have received the traffic model description parameters or figures from the customer for which the dimensioning campaign is performed. If the traffic model figures are not provided by the requesting customer, the dimensioning expert should request this data and clearly state that without exact TM the accuracy of the results of the capacity dimensioning campaign is reduced. Usually, a TM is provided in the form of a number of subscribers with a tr affic model per subscriber, as shown in Table 1 Exemplary TM: Se rvice

Usage per month

Usage in Busy Hour  

Voice

300 min

28,4 mErlang

WWW

10 MB

0,13 kbps

FTP

30 MB

0,39 kbps

Streaming

20 MB

0,26 kbps

Table 1

Exemplary TM

However, the availability of a customer-provided traffic model cannot always be taken for granted. In some cases, the dimensioning expert has no traffic model description from the customer which he can use as basis for capacity dimensioning. As a TM is a mandatory input for any capacity dimensioning campaign, a default TM should exist for the case that no other customer-specific traffic figures are available. As, usually, a TM is very operator- or, respectively, country-specific the default model shall be used only if operator data do no a llow for an estimation of expected traffic in the analyzed network. During dimensioning, the traffic model has a direct influence on the load that must be managed, all of the segments/interfaces; therefore, it must be clearly defined. Generally speaking, to avoid further problems, traffic modeling should be the first step when it comes to network dimensioning. The following workflow then can be applied:

10

Id:0900d80580973cca Confidential

DN0972027

LTE Traffic Model

Traffic dimensionig principles

Figure 4

Overall dimensioning workflow (E-UTRAN + EPC)

For detailed information about Radio, Access, or Core planning please refer to the corresponding guidelines.

2.1

Overbooking factor  The network has to be designed to provide sufficient capacity to its users. It is a compromise between costs and bandwidth. In practice, users transmit data for only a small portion of time. In fact the average throughput tends to be much lower than the requested subscription rate. The relation between subscription rate and average subscriber throughput in the busy hour (BH) is called overbooking factor (OBF). Subscription Rate OFB = ------------------------------------------------------------------------------ Average throughput within BH

Figure 5 

Subscription rate vs average throughput

In other words, this parameter gives information about the possibility to increase the number of users that can share the same resources. For typical dimensioning campaigns values for OBF are in the range of 10-25.

DN0972027

Id:0900d80580973cca Confidential

11

Traffic modeling

LTE Traffic Model

3 Traffic modeling The real traffic in a net work depends on the distribution of subscribers w ith individually varying traffic behavior using different terminal types over the planned area. This distribution is not constant, but more or less randomly varies over time. For dimensioning, however, usually it is assumed that only one TM is valid in the whole network (i.e. neither subscriber nor clutter differentiation). For big areas, it is possible to assume different traffic models for different clutters. In an LTE network, which is based on an 'all-IP' architecture, various applications can be served. One of the steps in the TM definition is to assume which application will be allowed for particular users. When different phases of traffic evolution are planned also a forecast of traffic increase in forthcoming years is essential. Figure 6 Steps in traffic modelling sums up all of the mentioned issues.

Figure 6 

12

Steps in traffic modelling

Id:0900d80580973ccc Confidential

DN0972027

LTE Traffic Model

Default LTE traffic model

4 Default LTE traffic model  As stated in the introduction, a default traffic model is required for those cases where a dimensioning expert needs to execute a dimensioning campaign, but did not receive a traffic model from the requesting customer. A default TM is defined by assuming a 'standard' subscriber who uses all accounted services in parallel. In this document, three different variants 'subscriber profile' have been defined (the exact difference between them are explained in more detail later), each of them characterized by a particular emphasis in the 'traffic mix':  •

Voice dominant

 •

Data dominant

 •

Mixed (voice + data)

The proposed TM follows the steps introduced in Traffic modeling. All corresponding assumptions have been collected below. 1.

The proposed TM assumes the usage of only one TM in the whole area. Users are uniformly distributed among all areas of different clutter types, and, within each area, the subscriber behavior is assumed to be the same.

2.

The amount of considered serv ices used by the network subscribers is limited to five (see table below).

3.

Each service is characterized by the mean holding time and transfer size: Service

Transfer size [kB]

Mean holding time [s]

VoIP

180

90

Video

720

90

Streaming (Live TV)

10500

600

Web browsing

500

600

FTP

5000

312

Table 2  4.

DN0972027

Service characteristics

 As indicated above, three different subscriber profiles were defined. Each profile is represented by different services usage. Three different traffic mix have been defined as shown in the Table 3 Traffic mix’s for different subscriber type.

Id:0900d80580973ccd Confidential

13

Default LTE traffic model

LTE Traffic Model

Unit

Value

1) Voice dominant subscriber profile Voice Usage per Subscriber

min

5

Video Usage per Subscriber

min

0

Streaming Usage per Subscriber

min

0

Web Usage per Subscriber

pages

0,333

FTP

kB

390,7

Data Usage per Subscriber

MB

1

2) Data dominant subscriber profile Voice Usage per Subscriber

min

0,1

Video Usage per Subscriber

min

0,1

Streaming Usage per Subscriber

min

2

Web Usage per Subscriber

pages

3,33

FTP

kB

7747

Data Usage per Subscriber

MB

10

3) Voice and data mixed profile Voice Usage per Subscriber

min

2,5

Video Usage per Subscriber

min

0,05

Streaming Usage per Subscriber

min

1

Web Usage per Subscriber

pages

1,665

FTP

kB

2913,5

Data Usage per Subscriber

MB

5

Table 3

Traffic mix’s for different subscriber type

 As the users'/subscribers' behavior is influenced by evolving operators' service policies and the extent of the users' f amiliarity with the new data services, the traffic mix of the different subscriber profile is expected to change in forthcoming years. Therefore, a further tuning of the TM and subscriber profiles will be necessary in future. 5.

For calculation of traffic demand, the number of Busy Hour Call Attempts ( BHCA) must be calculated. The Busy Hour Call Attempts per subscriber can be calculated using the following other formulas:  •

Voice, Video, and Streaming usage is expressed as 'minutes per month' , therefore the same formula applies to all of these. BHCA [#]

14

=

Id:0900d80580973ccd Confidential

Service usage per BH [min] --------------------------------------------------------------------------Servic e MHT

DN0972027

LTE Traffic Model

Default LTE traffic model

 •

For Web-browsing BHCA per subscriber, the following parameters are needed: BHCA [#]

 •

=

Web Usage per Subscriber  -----------------------------------------------------------------------------Tra ff ic Requests per session

Typical web session consists of 5 web pages; one page is equal to one Traffic Request .The average size of a web page is considered to be 100 KB. Considering this, the average data size of HTTP based traffic is 500 KB (5 web pages * 100 KB) as seen in the table. For calculating the File Transfer Protocol ( FTP ) BHCA, the following formula is used. BHCA [#]

=

Dat a volume ---------------------------------------------FTP transfer size

BHCA 1) Voice dominant subscriber profile VoIP

3,33

Video

0

Streaming (Live TV)

0

Web browsing

0,07

FTP

0,08

2) Data dominant subscriber profile VoIP

0,07

Video

0,07

Streaming (Live TV)

0,20

Web browsing

0,67

FTP

1,55

3) Voice and data mixed profile VoIP

1,67

Video

0,03

Streaming (Live TV)

0,10

Web browsing

0,33

FTP

0,58

Table 4

Calculated values for BHCA

If the application characteristics and BHCA of a system can be determined, then traffic per particular application for the busy hour can be obtained in the following way: Traffic_Volume_per_Application [kB]

DN0972027

=

BHCA ⋅ transfer_size_application

Id:0900d80580973ccd Confidential

15

Default LTE traffic model

LTE Traffic Model

Traffic volume per subscriber is the sum of all application volumes used by th is subscriber. For each area (and each phase), the traffic demand has to be accumulated for the number of subscribers who may use the specified services in this area and in this phase. The following definitions shall be used: OfferedTraffic [kB] 6.

=

number_of_subcarriers ⋅ Traffic_Volume_per_subscriber 

Traffic forecast plays a major role in cellular planning. It gives information about network growth for the years to com e. It allows to plan future network design, cost, geographical site distribution etc. The current prediction is that network traffic growth will be very fast. The estimated growth will be 15% per year. By using the following equation, the traffic evolution for the years 2009-2013 has been calculated. Tr aff ic Vo lu me ( 2009 + n )

=

1.15 ⋅ n ⋅ TrafficVolume (2009 )

n = 1, 2, 3...  A number of services have different requirements for uplink and downlink direction, e. g. download of music files requires high transfer capacity for downlink while the uplink is utilized only for short messages such as acknowledgements. Therefore, the calculations described below need to be done separately for the uplink and downlink. While the main reference document provides only values for downlink direction, the following simplified approach is used for the UL traffic volume calculation. The general idea is to introduce a parameter responsible for converting the DL volume to UL. This parameter is set to equal "4,3 : 1 " It is important to indicate that this is a simplified approach, and no differentiation on application basis has be en done - all applications are being treated with the same conditions.  All proposed values [kB during BH] have been collected in the following table. 2009

2010

2011

2012

2013

Voice dominant DL

UL

DL

UL

DL

UL

DL

UL

DL

UL

VoIP

600,00

139,53

690,00

160,47

793,50

184,53

912,53

212,22

1049,40

244,05

Video

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

Streaming (Live TV)

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

Web browsing

33,30

7,74

38,30

8,91

44,04

10,24

50,65

11,78

58,24

13,54

FTP

390,70

90,86

449,31

104,49

516,70

120,16

594,21

138,19

683,34

158,92

TOTAL

1024,00

238,14

1177,60

273,86

1354,24

314,94

1557,38

362,18

1790,98

416,51

2009

2010

2011

2012

2013

Data dominant

16

DL

UL

DL

UL

DL

UL

DL

UL

DL

UL

VoIP

12,00

2,79

13,80

3,21

15,87

3,69

18,25

4,24

20,99

4,88

Video

48,00

11,16

55,20

12,84

63,48

14,76

73,00

16,98

83,95

19,52

TOTAL

10240,00

2381,40

11776,00

2738,60

13542,40

3149,40

15573,76

3621,80

17909,82

4165,08

Id:0900d80580973ccd Confidential

DN0972027

LTE Traffic Model

Default LTE traffic model

Streaming (Live TV)

2100,00

488,37

2415,00

561,63

2777,25

645,87

3193,84

742,75

3672,91

854,17

Web browsing

333,00

77,44

382,95

89,06

440,39

102,42

506,45

117,78

582,42

135,45

FTP

7747,00

1801,63

8909,05

2071,87

10245,41

2382,65

11782,22

2740,05

13549,55

3151,06

TOTAL

10240,00

2381,40

11776,00

2738,60

13542,40

3149,40

15573,76

3621,80

17909,82

4165,08

2009

2010

2011

2012

2013

Voice + Data DL

UL

DL

UL

DL

UL

DL

UL

DL

UL

VoIP

300,00

69,77

345,00

80,23

396,75

92,27

456,26

106,11

524,70

122,02

Video

24,00

5,58

27,60

6,42

31,74

7,38

36,50

8,49

41,98

9,76

Streaming (Live TV)

1050,00

244,19

1207,50

280,81

1388,63

322,94

1596,92

371,38

1836,46

427,08

Web browsing

832,50

193,60

957,38

222,65

1100,98

256,04

1266,13

294,45

1456,05

338,62

FTP

2913,50

677,56

3350,53

779,19

3853,10

896,07

4431,07

1030,48

5095,73

1185,05

TOTAL

5120,00

1190,70

5888,00

1369,30

6771,20

1574,70

7786,88

1810,90

8954,91

2082,54

Table 5 

DN0972027

Traffic model for LTE

Id:0900d80580973ccd Confidential

17