Appendix-c: Fundamentals Of Radio Network Planning

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Contents 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 2 3 3.1 3.2 3.3 3.4 4 5 6 7

Mobile Radio Network Planning Tasks Collection of Basic Planning Data Terrain Data Acquisition Coarse Coverage Prediction Network Configuration Site Selection Field Measurements Tool Tuning Network Design Data Base Engineering Performance Evaluation and Optimization Repetition Radio Wave Propagation Path Loss Shadowing - Long Term Fading Multi Path Propagation - Short Term Fading Maximum Path Loss and Link Budget Cellular Networks and Frequency Allocation Traffic Models Exercises Solutions

MN1789EU07MN_0001

5 7 7 8 9 10 10 11 11 12 12 13 19 20 22 24 29 33 41 47 53

1

Siemens

2

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Objectives of Radio Network Planning To provide service to many subscribers

with high service quality

at low costs

Capacity for a traffic model Quality of service

Efficiency

l

service types

l

low blocking

l

low number of BS sites

l

call rate

l

low wait time

l

high frequency re-use

l

mobility

l

high speech quality

l

low call drop rate

Boundary conditions Physics:

frequency spectrum, radio propagation ® coverage & frequency planning

System:

receiver characteristics, transmit power

algorithms and parameter setting

channel configuration cell design & network structure link quality improvement

focal point of this course !

Fig. 1

MN1789EU07MN_0001

3

Siemens

Appendix-C: Fundamentals of Radio Network Planning

As shown in the figure below, the main topic of this course is adjustment of system parameters for the Siemens Base Station System (SBS) as part of the radio network planning process. Before going into the details of the system features and control parameters, this introduction chapter summarizes some basics on radio network planning: In the first and second section of this chapter the steps within the radio network planning process are explained. In sections 3 - 5 simple models concerning radio propagation, frequency re-use and teletraffic are presented. As each model they are only an approximation of reality. Nevertheless

4

l

they reflect the main physical effects,

l

they help to understand the meaning of parameters and the way of working the algorithms,

l

they allow to estimate parameter values.

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

1

Siemens

Mobile Radio Network Planning Tasks

MN1789EU07MN_0001

5

Siemens

Appendix-C: Fundamentals of Radio Network Planning

The mobile radio network is the connecting element between the mobile telephone users and the fixed network. In this network the base transceiver station equipment (BTSE) is the direct interface to the subscriber. It has to make radio communication channels available to the users and to care for a satisfactory signal quality within a certain area around the base station. This area may be split into different sectors (cells) which belong to one BTSE. Planning a mobile radio network is a complex task, because radio propagation along the earth surface is submitted to many influences due to the local environment. Furthermore the performance requirements to a radio network cover a wide field of applications which depend on the operators potentialities and goals. To respond to all these subjects, it is necessary to observe a certain sequence of tasks. The first step is to get knowledge about the customers/operators objectives and resources (basic planning data). On this basis it is possible to estimate the size of the project and to establish a coarse nominal cell plan. Then it is necessary to install a digital terrain data base into a planning tool which contains topo-graphical and morphological information about the planning region. This digital map permits to make more accurate predictions about the radio signal propagation as compared to the first rough estimation, and to create a more realistic cell structure, including the recommendable geographical positions of the base stations equipment(coarse coverage prediction). The network elements defined up to this moment have been found on a more or less theoretical basis. Now it has to be checked if the envisaged radio site locations may really be kept. A site survey campaign in accordance with the customer, who is responsible for the site acquisition, must clarify all problems concerning the infrastructure and technical as well as financial issues of the BTSE implementation. Inside a tolerable search area the optimum site meeting all these issues has to be selected. This site selection should also take into account particular properties of the area, e.g. big obstacles which are not recognizable in the digital maps. Field measurements, to be carried out in typical and in complex areas must give detailed informations about the radio characteristics of the planning region. The measurement results will then help to align the radio prediction tool for the actual type of land usage (tool tuning). Now, fixed site positions and an area-adapted tool being available, it is possible to start the detailed radio planning. The final network design has to care for both sufficient coverage and proper radio frequency assignment in respecting the traffic load and the interference requirements. The last planning step is the generation of a set of control parameters, necessary to maintain a communication while a subscriber is moving around. These parameters have to comply with the existing cell structure and the needs to handle the traffic load expected in each cell.

6

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

After commissioning of the network, the performance must be checked by the network operator by evaluation of statistical data collected in the operation and maintenance center. Situations of congestion or frequent call rejections may be treated by modification of the pertinent control parameters and lead to an optimized network. The individual planning steps are considered more closely in the following sections.

1.1

Collection of Basic Planning Data

The requirements of the network operator concerning traffic load and service area extension are basic data for the design of a mobile network . A coarse network structure complying with these requirements can be created on this basis. Two fundamental cell types are possible; their properties may be determined a)

by the maximum radio range of the involved transceiver stations and mobile terminals; the range is limited by the available transmit power and the noise figure of the receivers. This type is called a noise limited cell; it is typical for rural regions.

b)

or it may be determined by the limited traffic capacity of a cell in the case of high subscriber concentration. This leads to the implementation of small cells, mainly in urban areas where interference will become the major problem.

The result of this first planning step is a rough estimate of the network structure, called a nominal cell plan, which gives knowledge about the number of radio stations, their required technical equipment and their approximate geographical positions. Thus allowing to assess the monetary volume of the project.

1.2

Terrain Data Acquisition

Mobile communication occurs in a natural environment. The radio signal propagation is highly affected by the existing terrain properties like hills, forests, towns etc. Therefore the real mapping data must be taken into account by the planning tool. The signal level encountered by a subscriber in the street is influenced by absorbing, screening, reflecting and diffracting effects of the surrounding objects and along the radio path. To make realistic signal level predictions, the propagation models implemented in the prediction tool must be fed with the relevant terrain data.

MN1789EU07MN_0001

7

Siemens

Appendix-C: Fundamentals of Radio Network Planning

A very important factor for correct modeling is the morphographic classification of an area : l

building heights and density of built up areas (metropolitan, urban, suburban, village, industrial, residential) or forest, parks, open areas, water etc.

The screening by hills which may affect the coverage of a service area must be made evident by consideration of the terrain profile (height contour lines). The procurement of digital maps with these informations may be rather expensive. The prediction accuracy is directly related to the size of area elements (resolution) and to the reliability of these information (obsolescence of maps!)

1.3

Coarse Coverage Prediction

On the basis of the digital terrain data base and by using standard propagation models, which have been preselected to fit for special terrain types, it is possible to make field strength predictions without having a very detailed knowledge of the particular local conditions. By variation and modification of the site positions and antenna orientations, coverage predictions of rather good quality may be attained. Yet, the definitive site locations are subject to a later scheduled site selection process in accordance and by cooperation with the customer. The particular local characteristics must be introduced later by comprehensive survey measurements. These measurements will be used to upgrade the propagation models.

8

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

1.4

Siemens

Network Configuration

The results of the “coarse prediction“ steps will allow to define the radio network configuration and the layout of individual base stations. A first frequency allocation plan may also be derived from these predictions. The result might already be a well functioning network. But it is still based on assumptions. The actual impact of the natural environment must be considered in the following steps. Nevertheless, the “coarse planning“ results will help to better assess the special details brought in by the real situation. In designing the radio network one has to keep in mind the requirements emerging from an increasing subscriber number. A multiple phase implementation plan has to govern the network configuration concepts. In the initial phase a relatively low number of users has to be carried. On the other hand complete coverage of the service area has to be provided from the beginning. Existing sites of the first implementation phase must be useable in later phases. Increasing subscriber numbers (synonymous with increasing interference tendency!) should be responded by completion of the existing TRX-equipment and by addition of new sites. This means reconfiguration of the existing cell patterns and frequency reassignment. The planner should anticipate the future subscriber repartitions and concentrations from the beginning, in creating cell structures capable to respond to future needs. Increasing interference problems arising with higher site density may be overcome by downtilting of directional antennas initially mounted for maximum signal range, as now the radio cell areas will be smaller.

MN1789EU07MN_0001

9

Siemens

1.5

Appendix-C: Fundamentals of Radio Network Planning

Site Selection

The site positions found in the coarse planning process on a theoretical basis, must now be verified in a joint campaign, called site survey, between the customer and the radio network planner. All site candidates within a tolerable search area around the theoretical site positions must be checked. This check includes the availability of electric power and of data transmission lines. The most important topic is the possibility to install the antennas in a suitable height above the roofs or above ground. Environmental influences (screening obstacles, reflectors) have also to be regarded. The best fitting site should be selected. Another important task of this campaign is to declare a certain number of the radio sites be suitable to serve as „survey sites“. This means that radio field measurements shall be done with these stations as transmitters. The resulting measurements will be used for the alignment of radio propagation models. The environment of the survey sites should be typical for a considerable number of other radio sites.

1.6

Field Measurements

Digital terrain data bases (DTDB) as derived from topographical maps or satellite pictures do not contain all details and particularities of the existing environment. Especially in fast developing urban areas maps cannot keep pace with reality and thus reflect an obsolete status. Keeping maps on this quality level would be very expensive. The characteristics of built up zones and vegetation areas with respect to radio propagation differ in a wide range if we regard different countries. Even climatic conditions may influence the signal level. Knowledge about this specific behavior must be acquired by measurements. The survey measurements have to be carried out in typical areas. Evaluation of these measurements will result in models that can be applied in comparable areas as well. Special measurements must be carried out in very complex topographical regions where standardized propagation models will fail. The resulting models are valid exclusively for this measurement zone.

10

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

1.7

Siemens

Tool Tuning

The measurement results have to be compared with the predictions of proven standard models. The standard parameters will be slightly modified to achieve minimum discrepancies with the measurements, i.e. to keep the mean error and rmserror as low as possible. As the signal level is subject to statistical variations which cannot be predicted, the rms-error will never be zero. The reliability of the created models increases with the number of measurement runs that can be exploited. The new specific model may also be applied in other base stations located in similar environment.

1.8

Network Design

The area-specific models are the basis for the final planning steps. The detailed network design has to care for l

a suitable signal level throughout the planning area

l

sufficient traffic capacity according to the operators requirements

l

assignment of the pertinent number of RF-carriers to all cells

sufficient decoupling of frequency reuse cells to respect the interference requirements for co-channels and adjacent channels. Moreover, attention has to be paid to an optimized handover scenario in heavy traffic zones. The detailed planning process commits the final structure of the radio network and the configuration of the base stations. The capacity of digital data links connecting the radio stations to the fixed network elements may now be defined.

MN1789EU07MN_0001

11

Siemens

1.9

Appendix-C: Fundamentals of Radio Network Planning

Data Base Engineering

A cellular network is a living system with moving subscribers. The service must be maintained while mobiles change radio cells and superior organization units, called location areas. All control parameters, necessary to support this task, have to be administered and supervised in central data bases. There is a permanent signaling information exchange between mobiles, base stations and control centers. This signaling communication occurs on predefined time slots, called control channels which are assigned to one of the RF-carriers of each radio cell. Important control informations for each radio cell are : l

cell identification within the network

l

control carrier frequency

l

potential neighbor cells

l

minimum received signal level

l

maximum transmit power of a mobile

l

power reduction factor to perform power control

l

power margin for handover to neighbor cells

1.10

Performance Evaluation and Optimization

Regular performance checks must be carried out after commissioning of the network. These checks comprise the evaluation of statistical data collected in the “operations and maintenance center“ (OMC) as well as measurements by means of test mobile stations to explore e.g. handover events under realistic conditions; unwanted handover may lead to traffic congestions in certain cells, or may drain off traffic from other cells. Detection of multipath propagation problems caused by big reflecting objects is also subject to measurements. Another goal of these checks is to investigate the real traffic load and its distribution, as subscriber behavior in a living system will not necessarily reflect the original assumptions of the operator; assumed hot traffic spots may have been changed or shifted after a couple of years. Careful evaluation of the measurement data will help to optimize the network performance by modification of the system parameters. As the number of subscribers will normally increase in course of time, supervision and control of these parameters should become a permanent maintenance procedure.

12

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

2

Siemens

Repetition

MN1789EU07MN_0001

13

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Mobile Radio Network Planning Tasks l

Collection of basic planning data

l

Terrain data acquisition

l

Coarse coverage prediction

l

Network configuration

l

Site selection and field measurements

l

Tool tuning

l

Network design

l

Data base engineering

l

Performance evaluation and optimization

Collection of basic planning data

14

l

Customer must define basic network performance goals:

l

Size of service area and area types

l

Traffic load and distribution

l

Mobile classes and service quality

l

Future development (forecast)

l

Available RF - bandwidth

l

The resulting nominal cell plan is a first planning approach

l

to determine the required number of radio stations

l

to figure out the approximate equipment configuration

l

to get an idea of the financial volume of the project

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Terrain data acquisition Topographical and morphographical properties of the planning region must be compiled in a digital data base for further processing. Contents of the digital terrain data base DTDB: l

Height profile (topography)

l

Land coverage and usage (morphography)

Possible sources : l

Scanning of topographic maps

l

Processed satellite pictures or air pictures

Coarse coverage prediction A coarse coverage prediction based on the nominal cell plan and on the digital terrain data base: l

using standard propagation models

l

using standard antenna types

Results : l

Geographical distribution of the radio signal level

l

Coarse cell structure

l

Nominal position of the radio sites and antenna orientation

l

Search areas for final site positions

l

Knowledge about the attainable degree of signal quality

MN1789EU07MN_0001

15

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Network configuration Internal configuration of individual radio station: l

Equipment to be installed

Configuration of the radio network (network structure): l

Number of base station controllers BSC

l

Number of location areas

l

Definition of data lines between the network elements

Site selection and field measurements l

Selection of definitive radio site locations

l

Radio measurements in typical areas

l

Radio measurements in complex topographical regions

Tool tuning

16

l

Radio measurements are exploited to adapt standard propagation models to specific environmental conditions

l

Resulting models may be applied in similar environment

l

or are restricted to the special measurement area

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Network design The final radio planning is performed by means of the area - adapted models Planning goals: l

Sufficient signal level throughout the planning region

l

Sufficient traffic capacity according to subscriber distribution

l

Assignment of radio carriers to all cells

l

Low interference level for co-channels and adjacent channels

l

Definition of neighbor cells

Data base engineering Control and maintenance of the radio network requires parameters for l

Identification of serving cell and neighbor cells , i.e.: cell identity location area color code

l

Cell - allocated control- and traffic carriers

l

Maximum transmit power level

l

Minimum receive signal level

l

Power margin for handover to each neighbor cell

Performance evaluation and optimization l

By analyzing statistical data from maintenance center

l

Measurements performed by a test mobile station roaming about the operating radio network

MN1789EU07MN_0001

17

Siemens

18

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

3

Siemens

Radio Wave Propagation

MN1789EU07MN_0001

19

Siemens

Appendix-C: Fundamentals of Radio Network Planning

There are three main components of radio propagation which are discussed in the next section: l

mean path loss (loss due to the distance between MS-BS),

l

shadowing (long term fading),

l

multi path propagation ® short term (Raleigh) fading.

3.1

Path Loss

Standard path loss models are of the form: Lm[dB]= A + B log d [km] where Lm is the mean propagation path loss between the base station (BS) and the mobile station (MS) at a distance d. A:

unit loss at 1 km,

B:

propagation index or loss per decade.

The propagation coefficients A and B depend upon: l

the transmit frequency,

l

the MS and BS antenna heights,

l

the topography and morphology of the propagation area.

Examples are: 1. Free space loss: L0 = 32.4 + 20 log f [MHz] + 20 log d [km] or more important propagation in real environment - the famous Hata model: 2. Hata model The Hata model describes the mean propagation effects for large cells and distances d > 1 Km. For urban environment one has: A = 69.55 + 26.16 log f - 13.82 log Hb - a(Hm) B = 44.9 - 6.55 log Hb

20

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Frequency:

f [Mhz]

150...1000 -Mhz

BS antenna height:

Hb [m]

30...200 m

MS antenna height:

Hm [m]

a(Hm) = 0 for Hm = 1.5 m

Example:

Hm = 1.5 m

Hb = 50 m

®

A = 123.3

B = 33.8

f=900 Mhz

Path Loss for LargeCells - Hata Model (GSM 900) l

BS height 50 m

l

MS height 1.5 m

220 210 200 190 Suburban

Path Loss [dB]

180 Urban

170 Urban Indoor

160 150 140 130

Rural (quasi open)

120 Rural (open)

110 100 90 1

10

100

Cell radius [km]

Fig. 2

MN1789EU07MN_0001

21

Siemens

Appendix-C: Fundamentals of Radio Network Planning

For other environments (suburban, rural-quasi-open) the path loss per decade remains the same, but the unit loss is reduced by a certain amount. The free space loss and the Hata model are illustrated in the figure above. Models of this type are adequate for estimating the received level for large cells. However for a real network planning, refinements of the model and adaptations of parameters to morphological and topographical data and to measurement values are necessary (refer to section 1). The smaller the cells, the more important are the details of e.g. the building structure within the cell.

3.2

Shadowing - Long Term Fading

In larger cells where the BS antenna is installed above the roof top level, details of the environment near the MS are responsible for a variation of the received level around the mean level calculated by the models discussed above. Usually this variation of level - caused by obstacles near the MS (e.g. buildings or trees) - is described by the statistical model, i.e. the total path loss Ltot is given by the mean „distance“ path loss plus a random shadowing Ltot [dB] = Lm + S S<0: free line of sight, S>0: strong shadowing by e.g. a high building near the MS. S has a Gaussian distribution (see figure below) with mean value 0 and a standard deviation s which typically lies in the range s = 4...10 dB.

22

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

0.5 0.4 0.3 0.2 0.1 -3

-2

-1

0

Sh d

i

1

2

3

S/ [dB]

Fig. 3 Gaussian distribution of shadowing S

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

MN1789EU07MN_0001

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23

Siemens

Appendix-C: Fundamentals of Radio Network Planning

The length scale for variation of the long term fading is in the range 5 ... 100 m, i.e. the typical size of shadowing obstacles.

3.3

Multi Path Propagation - Short Term Fading

The superposition of several reflected waves arriving at the receiver on different paths and therefore with different amplitudes and phases causes peaks (constructive superposition) and deep fading dips (destructive superposition) of the received level. The length scale for variation (e.g. peak to peak) is given by the half of the transmission wave length, i.e. about 15 cm for GSM900 or 7.5 cm for DCS1800. An example for the variation of the received level due to short term fading is shown in the figure below. A comparison with the length scale for shadowing explains the names for these fading types. The statistics of the Raleigh fading is described in the following way: Consider the received level due the path loss and long term fading which is called local mean: LLOC[dBm]. The received local mean power is then given by Ploc[mW] = 10LlOC/10 Using this formula the probability density function for the received power P is given by: f(P) = 1/Ploc* exp(-P/Ploc) which means that the probability function for the signal amplitude P = A2 is given by a Raleigh distribution. Using these formulas and some mathematics, one can calculate the probability that the received level L (affected by Raleigh fading) is x dB below the local mean level Lloc: Prob (L - Lloc< x dB) = 1 - exp ( - 10 x/10)

24

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Example: x = 3 dB x = 0 dB x = -3 dB x = -6 dB x = -10 dB x = -20 dB

Prob = 86,5 % Prob = 63,0 % Prob = 39,5 % Prob = 22,0 % Prob = 9,5 % Prob = 1,0 %

Changing the transmission frequency and therefore the wave length, changes the position of Raleigh peaks and dips. This means that at a given position, the received level affected by Raleigh fading in general differs for different transmission frequencies. The higher the frequency difference the lower is the correlation for the receive signal for the different frequencies. The coherence bandwidth Bcoh is defined as the frequency difference at which this correlation has decreased to 0.5. The coherence bandwidth depends upon the spread of arrival times of the different multi path components of the received signal. This spread is called delay spread DT:

Coherence Bandwidth and Delay Spread Bcoh =

1 2pDT

i.e. the higher the delay spread the lower is the coherence bandwidth. The delay spread depends upon the propagation environment. Typical values are: l

10 µs for hilly terrain (corresponding to path length between difference of 3 km).

l

0.1 ... 1 µs for urban area (corresponding to path length between difference of 30 ... 300 m).

Keeping in mind that a Raleigh fading dip of more than 10 dB occurs with a probability of 10 %, measures should be provided to combat Raleigh fading:

Means to combat Raleigh fading: l

Averaging of Raleigh fading over speech frames (interleaving of 8 bursts) Frequency Hopping spacing between frequencies in hopping sequence >> coherence bandwidth Motion (speed v) Example: v=50 km/h, distance between bursts = TDMA frame length T = 4.6 ms ® distance between MS positions at subsequent bursts D = 6.4 cm ® distance for 8 bursts_ 8 * D ~ 50 cm > 3 * wavelength

l

Combining of signals received at positions of mutually uncorrelated fading Antenna Diversity spacing between RX antennas >> half wavelength

MN1789EU07MN_0001

25

Siemens

26

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Fig. 4

MN1789EU07MN_0001

27

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Short Term Fading

Fig. 5

28

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

3.4

Maximum Path Loss and Link Budget

The maximum radius of a cell depends on the maximum possible path loss between transmitter and receiver, i.e. upon the difference between the maximum output power level EIRP (emitted isotropic radiation power) at the transmitter antenna and the required input power level (RIPL) at the receiver antenna. Output BTS: EIRPBTS = Power Amplifier Output - Combiner Loss - Downlink Cable Loss + Antenna Gain Power Amplifier Output:

25 Watt = 44 dBm (GSM900) (higher power amplifier output power in further BTS versions)

Combiner Loss Combiner Type

1:1

2:1

4:1

Duplexer

2.7 dB

2.7 dB

5.9 dB

Hybrid Combiner

2.0 dB

5.2 dB

8.4 dB

The ratio x:1 denotes the number of carriers which are combined. In the case of hybrid combiners the signals are fed to 1 transmitter antenna. In the case of duplexers the signals are fed to 2 antennas (on air combining) which are used for transmission as well as for reception. Using these antennas for reception, a two branch (maximum ratio) antenna diversity combining can be realized. This means that - using Duplexers - two antennas per cell are needed, whereas when using Hybrid Combiners and applying Antenna Diversity two receive plus one transmit antenna is needed. Downlink Antenna Cable Loss: 3 dB (example) Antenna Gain (example):

MN1789EU07MN_0001

16 dB (typical value for 600 half power beam width antenna)

29

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Output MS: For the MS there is no need combining different carriers; and the cable loss and antenna gain reduce to zero. The EIRP depends upon the power class of the MS specified in GSM Rec 05.05: Power Class (GSM 05.05)

Max. Output Power (GSM900)

Max. Output Power (DCS1800)

1

--

1 Watt = 30 dBm

2

8 Watt = 39 dBm

0.25W = 24 dBm

3

5 Watt = 37 dBm

4 Watt = 36 dBm

4

2 Watt = 33 dBm

5

0.8 Watt = 29 dBm

Input BTS: The required input power level RIPL at the BTS antenna is given by RIPLBTS = Receiver Sensitivity Level - Antenna Diversity Gain + Uplink Cable Loss - Antenna Gain Receiver Sensitivity Level < - 104 dBm The receiver sensitivity level is defined in GSM Rec. 05.05 for scenarios where short term Raleigh fading is (at least) partly averaged either by motion or by frequency hopping. The receiver sensitivity level has been measured to be better than required by GSM Rec. 05.05. Antenna Diversity Gain:

4 dB (for a typical scenario).

The gain which can be achieved by antenna diversity strongly depends upon the propagation environment, the velocity of the mobile and on whether frequency hopping is applied or not. For a typical urban environment, a mobile speed of 3 km/h and frequency hopping applied the antenna diversity gain is about 4 dB. Uplink Cable Loss

3 dB

without tower mounted preamplifier RXAMOD

0 dB

with tower mounted preamplifier RXAMOD

The (uplink) cable loss from the antenna to the receiver input can be compensated using a tower mounted amplifier called RXAMOD. It should be noted that this preamplifier cannot be used together with on air combining (Duplexers). Antenna Gain (example):

30

16 dB (typical value for 600 half power beam width antenna)

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Input MS: For the MS there is neither antenna gain nor antenna diversity gain. Cable losses can be neglected. Therefore the required input power level at the MS antenna is given by the MS receiver limit sensitivity as specified by GSM 05.05: l

104 dBm for class 2 and 3 (GSM900),

l

102 dBm for class 4 and 5 (GSM900),

l

100 dBm for class 1 and 2 (DCS1800)

Maximum allowed path loss (Link Budget) downlink

Ld[dB] = EIRPBTS - RIPLMS

uplink

Lu[dB] = EIRPMS - RIPLBTS

Example: Duplexers 2:1:

® no RXAMOD, uplink cable loss = 3 dB

MS of Power Class 3:

® EIRPMS= 37 dBm

Antenna Diversity Gain:

4 dB

® Lu[dB] = 37 dBm - (- 104 dBm - 16 dBi + 3 dB - 4 dB) = 158 dB ® Ld[dB] = 44 dBm - 3 dB - 3 dB + 16 dBi - (- 104 dBm) = 158 dB l

i.e. there is a symmetric link budget for uplink and downlink.

l

Requirement: Area Coverage Probability: 90 % ¨Coverage Probability at Cell Border: 75 %

l

Standard Deviation of Shadowing: s= 6 dB ® 75 % value of Shadowing: S75%= 4 dB

l

allowed loss L - S75% = 154 dB

® Lm = L - S75% = 154 dB Path loss model (Hata):

Lm [dB] = 123.3 + 33.7 log d [km] ® Cell Radius: dmax =10 (154-123.3)/33.7 = 8.15 km

Example 2: Designing a radio cell for mainly MS of Power Class 4 (instead of power class 3), the following values for link budget are obtained: Lu[dB] = 154 dB Ld [dB] = 156 dB To obtain a symmetric link budget, the power amplifier output power of the BTS has to be reduced by 2 dB. This is done using the O&M parameter BS_TXPWR_RED:

MN1789EU07MN_0001

31

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Object

DB Name

Range

Meaning

TRX

PWRRED

0, 1, ...6 * 2dB

Reduction of BTS power amplifier output

Reducing the BTS output power has the advantage that less downlink interference is caused by this cell. If there are also some mobiles of Power Class 2 and 3 within the cell designed for mobiles of Power Class 4, their maximum transmit power has to be limited for a link budget balance. This is the reason behind the following parameters: Specification Name

DB Name/ Object

Range

Meaning

MS_TXPWR_MA X

MSTXPMAX / BTS-B

2...15 GSM 0...15 DCS * 2 dB

Maximum TXPWR a MS is allowed to use on a dedicated channel (TCH or SDCCH) in the serving cell GSM: 2 = 39 dBm, 15 = 13 dBm DCS: 0 = 30 dBm, 15 = 0 dBm PCS: 0 = 30 dBm, 15 = 0 dBm 30 = 33 dBm, 31 = 32 dBm

0...31 * 2 dB

Maximum TXPWR a MS is allowed to use on the uplink common control channel (Random Access Channel, RACH) in the serving cell: GSM: 0 = 43 dBm,19 = 5 dBm DCS: 0 = 30 dBm, 15 = 0 dBm

MS_TXPWR_MA MSTXPMAXCH / X_CCH BTS-C

Another effect illustrated by this example is the following: Since there is a balanced link budget Lu[dB] = Ld[dB], but a difference of the receiver sensitivity level for the MS and BTS of 2 dB, there is difference between the mean downlink and uplink received level RXLEV of about 2 dB: RXLEV_DL - RXLEV_UL ~ 2 dB. The consequence is that level threshold for e.g. the handover algorithm have to be set 2 dB higher for the downlink than for the uplink.

32

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

4

Siemens

Cellular Networks and Frequency Allocation

MN1789EU07MN_0001

33

Siemens

Appendix-C: Fundamentals of Radio Network Planning

One important characteristic of cellular networks is the re-use of frequencies in different cells. By re-using frequencies, a high capacity can be achieved. However, the re-use distance has to be high enough, so that the interference caused by subscribers using the same frequency (or an adjacent frequency) in another cell is sufficiently low.

34

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

MS

Siemens

Co-channel Re-us e Cells

Interferer Carrier

Re-us e Ditance D

Cell Radius R

Fig. 6

MN1789EU07MN_0001

35

Siemens

Appendix-C: Fundamentals of Radio Network Planning

To guarantee an appropriate speech quality, the carrier-to-interference-power-ratio CIR has to exceed a certain threshold CIRmin which is 9 dB for the GSM System (GSM Rec. 05.05). taking the situation of the example above and a path loss model L = A + B log d, one has C/Itot[Watt] = C / (I1 + ... + INI) ~ C / (NI * I1)

NI: number of interferes

or in dB C/Itot [dB]

= C[dB] - Itot[dB] ~ B log D - B log R - 10 log NI = B log D/R - 10 log NI > CIRmin + LTFM (x%)

By introducing the long term fading margin LTFM (x%) for a required coverage probability of x%, the effect of shadowing is taken into account. For homogeneous hexagonal networks frequencies can be allocated to cells in a symmetric way. Defining the cluster size K as group of cells in which each frequency is used exactly once, the following relations between Cluster Size, Cell Radius and Re-use Distance are obtained.

36

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Frequency Re-use and Cluster Size

m

D

D R

n

r

Fig. 7

Outer Cell Radius

l

R

Inner Cell Radius

l

r = 0.5 x 3 x R

Re-use Distance

l

D = R x 3 x (n 2 +m 2 +nm)

D = 3xK R

Cluster Size:

Group of cells in which each frequency is used exactly once K = (n 2 + m 2 + nm) n, m = 0, 1, 2, 3, ... K = 1, 3, 4, 7, 9, 12, 16, 19, ...

MN1789EU07MN_0001

37

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Inserting the formula for the cluster size into the formula for the minimum CIR one obtains: 0.5 * B log 3 K > CIRmin + LTFM (x%) + 10 log NI which gives a lower bound for the cluster size which can be used. For a given cluster size K and total number of frequencies Ntot, the number of frequencies per cell Ncell is given by: Ncell = Ntot/K i.e. the capacity of a cell can be increased by reducing the cluster size. A reduction of cluster size can be achieved by l

reducing the number of interferes ® Sectorisation.

l

reducing the interference from co-channel cells ® Power Control, Discontinued Transmission, ...

Examples for sectored network structure are shown in the figures below. Methods for interference reduction are discussed in chapter 6. Obviously a real network does not have such a regular hexagonal structure and frequency allocation is performed by planning tools using complex algorithms for optimizing the CIR in each cell. The objective is to achieve a high mean value of frequencies per cell . The ratio = Ntot/Ncell can viewed as the mean cluster size in such an inhomogeneous environment. The capacity of the radio network depends upon the available number N of radio channels per area F (e.g. F = 1 km2).

N N N N 1 1 = Ncell x BTS = CPF x tot x = CPF x tot x F F K F / NBTS K CA

38

NBTS:

number of BTS

CA:

cell area

CPF:

channel per frequencies

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Omnicells - Cluster 7

7 6 7 6

2 1 5

2 1 5 3 4 7 6

3 4 7 6 2 1 5

7 6 2 1 5 3 4

2 1 5 3 4 7 6

3 4 7 6 2 1 5

2 1 5

3 4

3 4

Fig. 8 Example for homogeneous frequency allocation

MN1789EU07MN_0001

39

Siemens

Appendix-C: Fundamentals of Radio Network Planning

3-Sector Cloverleaf - Cluster 3 x 3

2a

2a 1a 1c

2c

1a 1c

3a

1b 2a

2c

3c

3c

1c

2c

3c

1a

2b 1c 3b

1c 3b

2a

3b 1a

2b

3c

3a

1b

3c

3a

1b

2b 3a

1b

2c

2a

3b

2c

2a

3b

1c

1a

1c

1a

2b 3a

1b

1a

2b

2c 3a

1b 2a

2c

2b

3c

3b

2b 3a

1b 3c

3b

Fig. 9 Example for homogeneous frequency allocation

40

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

5

Siemens

Traffic Models

MN1789EU07MN_0001

41

Siemens

Appendix-C: Fundamentals of Radio Network Planning

A traffic model reflects the behavior of the subscribers, as their mobility, the mean call rate or call duration. It is needed e.g. for calculating the required total number of channels within a cell and how to split them between traffic and control channels. These traffic model information is always a mixture between field observations in similar networks and arbitrary assumptions. Traffic data are variable in time, therefore statistical characterization is used. The goal of planning is to manage traffic even in busy hour. In mobile networks we have to evaluate two main factors: l

user mobility

l

communications

User mobility: The user moves with a velocity v. For example the handover and location update rates depend on this velocity. Communications: The number of subscriber in a cell, the traffic per subscriber has to be considered. Furthermore, one needs information the mean call duration, the mean call cell rate (or busy hour call attempt BHCA). separately for mobile originating calls (MOC) and mobile terminating calls (MTC).

42

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

An example for a traffic model is given in the table below: number of call attempts (MOC+MTC) per subscriber per hour

1,1

percentage of MOC

58 %

percentage of ‘engaged’ in the case of an MOC

19,8 %

duration of TCH occupation in the engaged case

3s

no answer from a person called by MOC

14,4 %

mean TCH occupation for this case

30 s

percentage of successful MOC

65,8 %

mean time for ringing (MOC)

15 s

percentage of MTC

42 %

no paging response

32,5 %

duration of TCH occupation in this case

0s

no answer from a mobile subscriber

13,5%

means TCH occupation fir this case

30 s

successful MTC

54,0 %

mean time for ringing (MTC)

5s

mean call duration (MOC/MTC)

115 s

mean TCH occupation call attempt

83 s

TCH load per subscriber

0,025 Erl

time for MOC/MTC setup signaling on SDCCH (authentications, ...)

3s

time for a location update

5s

number of location update per subscriber per hour

2,2

resulting SDDCCH load per subscriber (no TCH queuing applied)

0,004 Erl

Standard traffic model for GSM

The formula for calculating the load on the respective dedicated channel are given on the next page.

MN1789EU07MN_0001

43

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Load on Dedicated Channels SDCCH load [Erl]:

SUBSCR *

((MTC_PR_ph + MOC_ph) * T_SETUP + LU_ph * T_LU+ IMSI_ph * T_IMSI + SMS_ph * T_SMS)

TCH load [Erl]:

SUBSCR *

(MTC_PR_ph + MOC_ph) * T_CALL

SUBSCR:

number of subscribers within the cell

MTC_PR_ph:

mobile terminating calls per subscriber per hour with paging response

MOC_ph:

mobile terminating calls per subscriber per hour

LU_ph:

location updates per subscriber per hour

IMSI_ph:

IMSI attach/detach per subscriber per hour

SMS_ph

short message service per hour

T_SETUP:

mean time [sec] for call setup signaling on SDCCH

T_LU:

mean time [sec] for location update signaling

T_IMSI:

mean time [sec] for IMSI attach/detach signaling on SDCCH

T_SMS:

mean time [sec] for short message service

T_Call:

mean TCH occupation time per call

For the values of the traffic model above one has TCH load per subscriber: SDCCH load per subscriber:

44

25 mErl 4 mErl

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

p=1%

p=3%

p=5%

p=7%

0.01 0.15 0.46 0.87 1.36 1.91 2.50 3.13 3.78 4.46 5.16 5.88 6.61 7.35 8.11 8.88 9.65 10.44 11.23 12.03 12.84 13.65 14.47 15.29 16.13 16.96 17.80 18.64 19.49 20.34 21.19 22.05 22.91 23.77 24.64 25.51 26.38 27.25 28.13 29.01 29.89 30.77 31.66 32.54 33.43 34.32 35.22 36.11 37.00 37.90

0.03 0.28 0.72 1.26 1.88 2.54 3.25 3.99 4.75 5.53 6.33 7.14 7.97 8.80 9.65 10.51 11.37 12.24 13.11 14.00 14.89 15.78 16.68 17.58 18.48 19.39 20.31 21.22 22.14 23.06 23.99 24.91 25.84 26.78 27.71 28.65 29.59 30.53 31.47 32.41 33.36 34.30 35.25 36.20 37.17 38.11 39.06 40.02 40.98 41.93

0.05 0.38 0.90 1.53 2.22 2.96 3.74 4.54 5.37 6.22 7.08 7.95 8.84 9.37 10.63 11.54 12.46 13.39 14.31 15.25 16.19 17.13 18.08 19.03 19.99 20.94 21.90 22.87 23.83 24.80 25.77 26.75 27.72 28.70 29.68 30.66 31.64 32.62 33.61 34.60 35.58 36.57 37.57 38.56 39.55 40.54 41.54 42.54 43.53 44.53

0.08 0.47 1.06 1.75 2.50 3.30 4.14 5.00 5.88 6.78 7.69 8.61 9.54 10.48 11.43 12.39 13.35 14.32 15.29 16.27 17.25 18.24 19.23 20.22 21.21 22.21 23.21 24.22 25.22 26.23 27.24 28.25 29.26 30.28 31.29 32.31 33.33 34.35 35.37 36.40 37.42 38.45 39.47 40.50 41.53 42.56 43.59 44.62 45.65 46.69

n 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100

p=1%

p=3%

p=5%

p=7%

38.80 39.70 40.60 41.50 42.41 43.31 44.22 45.13 46.04 46.95 47.86 48.77 49.69 50.60 51.52 52.44 53.35 54.27 55.19 56.11 57.03 57.96 58.88 59.80 60.73 61.65 62.58 63.51 64.43 65.36 66.29 67.22 68.15 69.08 70.02 70.95 71.88 72.81 73.75 74.68 75.62 76.56 77.49 78.43 79.37 80.31 81.24 82.18 83.12 84.06

42.89 43.85 44.81 45.78 46.74 47.70 48.67 49.63 50.60 51.57 52.54 53.51 54.48 55.45 56.42 57.39 58.37 59.34 60.32 61.29 62.27 63.24 64.22 65.20 66.18 67.16 68.14 69.12 70.10 71.08 72.06 73.04 74.02 75.01 75.99 76.97 77.96 78.94 79.93 80.91 81.90 82.89 83.87 84.86 85.85 86.84 87.83 88.82 89.80 90.79

45.53 46.53 47.53 48.54 46.54 50.54 51.55 52.55 53.56 54.57 55.57 56.58 57.59 58.60 59.61 60.62 61.63 62.64 63.65 64.67 65.68 66.69 67.71 68.72 69.74 70.75 71.77 72.79 73.80 74.82 75.84 76.86 77.87 78.89 79.91 80.93 81.95 82.97 83.99 85.01 86.04 87.06 88.08 89.10 90.12 91.15 92.17 93.19 94.22 95.24

47.72 48.76 49.79 50.83 51.86 52.90 53.94 54.98 56.02 57.06 58.10 59.14 60.18 61.22 62.27 63.31 64.35 65.40 66.44 67.49 68.53 69.58 70.62 71.67 72.72 73.77 74.81 75.86 76.91 77.96 79.01 80.06 81.11 82.16 83.21 84.26 85.31 86.36 87.41 88.46 89.52 90.57 91.62 92.67 93.73 94.78 95.83 9689 97.94 98.99

Erlang B formula

MN1789EU07MN_0001

45

Siemens

46

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

6

Siemens

Exercises

MN1789EU07MN_0001

47

Siemens

48

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Exercise Title:

Calculation Loss / Gain

Task æ P ö LP = 10 log ç -3 ÷ dBm è 10 ø

reference P = 1 mW

æ U ö LU = 20 log ç -6 ÷ dBmV è 10 ø

reference U = 1 µV

æ P ö Loss A = 10 log ç in ÷ dB è Pout ø æP ö Gain G = 10 log ç out ÷ dB è Pin ø P=

U2 R

æ U ö LU = 20 log ç ÷ dBU è 0,775 ø

reference = 775 mV, 600 9

æ U ö A = 20 log ç in ÷ è Uout ø æU ö G = 20 log ç out ÷ è Uin ø

1. Amplifier: 100 mVin, 1 Vout. Calculate the gain 2. Amplifier: 2 mWin, 5 Wout. Calculate the gain 3. Amplifier: 20 dBmin, two steps amplification with 7 dB, 3 dB gain. Calculate the gain.

MN1789EU07MN_0001

49

Siemens

Appendix-C: Fundamentals of Radio Network Planning

Appendix Exercise 1 Power classes for MS/BTS Class

Watt

dBm

1 2 3 4 5

20 8 5 2 0,8

43 39 37 33 29

ü ï ý MS ï þ

1 2 3 4 5 6 7 8

320 160 80 40 20 10 5 2,5

55 52 49 46 43 40 37 34

ü ï ï ý BTS ï ï þ

Conversion dBm « Watt Watt 4·

10 10 10 10 10 10

dBm -14

-104 -20 -10 0 10 20 30 33 44 47 50

-5 -4 -3 -2 -1

1 2 25 50 100 Maximum Range Example: SBS, GSM Power Amplifier Watt Combiner 2:1 Cable Antenna gain

25

^

44 - 8 - 3 +18

dBm dB dB dB

51

dBm ^ 125 Watt

Sending power

50

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Handy sensitivity

-102 dBm

Þ possible loss

153 dBm

153 dBm

Siemens

^ 8 km free area 3 km urban area 1 km downtown

Typical loss values: Fading: Glass: Wall: Shopping Mall: House

MN1789EU07MN_0001

6 dB 5 dB 12 dB 25 dB 15 dB

51

Siemens

52

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

7

Siemens

Solutions

MN1789EU07MN_0001

53

Siemens

54

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001

Appendix-C: Fundamentals of Radio Network Planning

Siemens

Solution Title:

Calculation Loss / Gain

Task

1.

æ 1ö G = 10 log ç ÷ = 20 dB è 0,1ø

2.

æ 5 ö G = 10 log ç ÷ = 34 dB è 0,002 ø

3. Power out = 20 + 7 + 3 = 30 dBm 30 dBm ^ 1 Watt

MN1789EU07MN_0001

55

Siemens

56

Appendix-C: Fundamentals of Radio Network Planning

MN1789EU07MN_0001