11. Cellular Planning and Capacity

| University of Dar es Salaam | Department of Electronics and Telecommunications Engineering |

_______________________________________________________________________________

Cellular Planning and Capacity

------------------------------------------------------------------------------------------------------------------------

| TE 412 Introduction to Wireless Communications | Christine Mwase | 10/03/2014 |

Cellular Planning and Capacity


_______________________________________________________________________________

Cellular Planning

We want

• complete area coverage

• no dead spots

For economic reasons, hexagons are normally assumed

Use

polygons

------------------------------------------------------------------------------------------------------------------------


For economic reasons, hexagons are normally assumed (result in more efficient design than squares or triangles)

Rx 3=


_______________________________________________________________________________

------------------------------------------------------------------------------------------------------------------------



_______________________________________________________________________________

Frequency Planning and Re-use

• An operator is allocated a certain bandwidth allocation.

• Bandwidth is sub-divided into sub-bands.

– Exact number of sub-bands is determined by

• the interference tolerate of the radio system and

• the type of basestation antenna arrangement employed.

------------------------------------------------------------------------------------------------------------------------


• the type of basestation antenna arrangement employed.

• The number of sub-bands is referred to as the cluster size

– Cells are planned in a cluster using entire frequency allocation.

– The previous figure assumed a cluster size of four

• Total frequency allocated is divided into four sub-bands and assigned to the individual cells.


_______________________________________________________________________________

------------------------------------------------------------------------------------------------------------------------


Hexagonal Axes and Co-ordinate System


_______________________________________________________________________________

Normalised Distance to cell centre

• The most convenient co-ordinate system for a hexagonal cellular structure are axes inclined at a 60 degree angle.

• Assuming origin is at (0,0) and restricting u and v to integer values (i,j), the distance to the cell centre is:

------------------------------------------------------------------------------------------------------------------------


values (i,j), the distance to the cell centre is:

• The normalised distance between any adjacent cell sites is unity (i=1, j=0 or i=0, j=1).

( ) ijjiDn −+=2


_______________________________________________________________________________

To find the nearest co-channel cell:

• Move perpendicular to any of

the six surfaces of the original

cell, pass i=4 cells, rotate 60

degrees counter-clockwise,

move along j=2 cells. (red)

------------------------------------------------------------------------------------------------------------------------


move along j=2 cells. (red)

• Move perpendicular to any of

the six surfaces of the original

cell, pass j=2 cells, rotate 60

degrees clockwise, move along

i=4 cells. (blue)


_______________________________________________________________________________

Cellular Geometry and Area

------------------------------------------------------------------------------------------------------------------------



_______________________________________________________________________________

Cluster Size• Cells designated A are the six nearest cochannel cells of the

centre cell. They are at the vertices of the larger hexagonal cell of radius D. The radius of the larger cell, D, is given by:

( )ijjiRD ++=2222

3

------------------------------------------------------------------------------------------------------------------------


• Since the area of a hexagon is proportional to the square of its radius, the area of the larger hexagon is given by:

where k represents a fixed constant

( )ijjiRD ++= 3

( )( )ijjiRkA el ++=222

arg 3


_______________________________________________________________________________

Cluster Size

• Similarly, the area of the small hexagon is given by:

( )2RkAsmall =

------------------------------------------------------------------------------------------------------------------------


• Therefore, the ratio of the large hexagonal area to the small hexagonal area is given by:

( )small

( )ijjiAA smallel ++=22

arg 3


_______________________________________________________________________________

Cluster Size

• The larger hexagon has total of 3N enclosed cells.

– the N shaded centre cells (N = 7)

– one third of the six surrounding N cell clusters.

------------------------------------------------------------------------------------------------------------------------


• Since the enclosed area is proportional to the number of cells we can write:

( )( ) ( )( ) ( )ijjiNhenceijjiRkNkA

RkkhencekA

el

small

++=++==

==

222222arg

222

33.

1.


_______________________________________________________________________________

Cluster Size

• Since i and j are restricted to integer values, cluster size N is restricted to limited set of discontinuous values. It is not possible to plan a system with N = 2, 5, 6, 8 … etc.

• Combining the above equation for D and N we can show that:

------------------------------------------------------------------------------------------------------------------------


• Combining the above equation for D and N we can show that:

• This relates the cell radius and reuse distance to the cluster size. The term D/R is referred to as the cochannel reuse ratio.

NR

D 3=


_______________________________________________________________________________

Cluster size as a function of i and j

i j N

1 0 1

1 1 3

------------------------------------------------------------------------------------------------------------------------


2 0 4

2 1 7

3 0 9

3 1 13

4 0 16


_______________________________________________________________________________

Cluster size as

a function of

Co-channel

Interference

------------------------------------------------------------------------------------------------------------------------



_______________________________________________________________________________

Cluster size as a function of Cochannel Interference

• System can be modelled based on the wanted signal and the interference from the surrounding ‘first tier’ cochannel cells.

• Worst case scenario: user on cell boundary (distance R from basestation). The first tier interfering cells lie at distance D-R.

------------------------------------------------------------------------------------------------------------------------


• The signal power from the centre basestation is inversely proportional to the n-th power of the distance– n = propagation path loss index (n = 2 for LOS communications):

nR

S1

α


_______________________________________________________________________________


• The interference power from each of the first tier cells is inversely proportional to the n-th power of the distance (D-R):

nD

I1

1α

------------------------------------------------------------------------------------------------------------------------


• Since there are six interfering cells (assuming omni antenna), actual interference is six times that of 1. Resulting S/I ratio is:

nD

( )( )[ ] n

n

ISRDhenceRD

I

S 16

6==


_______________________________________________________________________________


( )[ ] nISN2

31 6=

------------------------------------------------------------------------------------------------------------------------


Thus, N can be found from the S/I ratio needed by the system.

– GSM and DCS-1800: an S/I of 9 dB is required• hence a reuse cluster size of 4.39 (path loss exponent assumed to be 3).

– TACS: an S/I of 17 dB is required• hence a reuse cluster size of 14.9.

– Based on this alone, GSM appears to offer over three times the capacity of TACS. In practice many other factors can affect system capacity.


_______________________________________________________________________________

------------------------------------------------------------------------------------------------------------------------



_______________________________________________________________________________

Capacity Evaluation

Capacity = Function [total bandwidth, channel bandwidth, cluster size, cell area]

• Channel bandwidth = average bandwidth needed to support a single communications channel.

– For TDMA systems, divide carrier bandwidth by number of user slots.

------------------------------------------------------------------------------------------------------------------------


• Cluster size is a function of C/I tolerance and sectorisation strategy.

• Since capacity is often quoted ‘per km squared’, the average cell radius is also used in the formulation.

2// kmMHzErlangs

bandwidthTotalareaCellsizeCluster

channelsofNumberTotalCapacity

××=


_______________________________________________________________________________

Capacity Evaluation

• Channel bandwidth varies depending on technology.

– TACS , AMPS (FM based) used ≈ 25 – 30 kHz per voice channel.

– GSM support 8 or 16 voice channels in 200 kHz band � average channel bandwidths of ≈ 25 kHz and 12.5 kHz.

– In newer analogue systems, bandwidth reduced to 12.5 kHz.

------------------------------------------------------------------------------------------------------------------------


– There is interest in reducing the required bandwidth still further.

• Channel bandwidth is reduced at expense of C/I performance. Total number of calls is directly related to channel bandwidth. However, more bandwidth efficient schemes often need larger cluster sizes and reuse distances due to their greater sensitivity to cochannel interference.


_______________________________________________________________________________

Capacity Evaluation• Capacity is also a function of cell area.

– higher capacity can be obtained with smaller cell sizes. These

microcells aim to offer high subscriber densities. They offer high

capacity (in terms of subscribers / MHz / km square), but many

basestations are needed for continuous wide coverage.

------------------------------------------------------------------------------------------------------------------------


basestations are needed for continuous wide coverage.

• To determine number of subscribers that can be supported,

average traffic density per user can be included in equation.

• Average subscriber creates ≈ 0.02 Erlangs of traffic. Hence:

2// kmMHzssubscriberusersErlangsbandwidthTotalareaCellsizeCluster

channelsofNumberTotalCapacity

×××=


_______________________________________________________________________________

Example

A digital cellular system has a total bandwidth of 15 MHz and operates with a re-use factor of 4 and a cell radius of 2 km. The network comprises 20 basestations and each 200 kHz carrier is made up of 8 TDMA voice channels. Assume that each user represents a traffic load of 0.02 Erlangs.

------------------------------------------------------------------------------------------------------------------------


each user represents a traffic load of 0.02 Erlangs.

– Calculate the total subscriber capacity in each cell.

– Next determine the total capacity in terms of subscribers per MHz per kilometre square.

– Finally, what is the total subscriber capacity of the network?


_______________________________________________________________________________

ExamplePart 1:

• Calls per cell = ( 15 / ( ( 0.200 / 8 ) × 4) = 150

• Subscribers per cell = 150 / 0.02 = 7500

------------------------------------------------------------------------------------------------------------------------


Part 2:

• Assuming 200kHz call bandwidth, 15MHz supports 600 calls.

• Channels are used over cluster area 4 × 12.6 = 50.4

• � Total capacity = 600 / (15 × 50.4) = 0.794 calls / MHz /

• Subscriber capacity = (0.794/0.02) = 39.68 subscribers/MHz/

2km

2km

2km

2km


_______________________________________________________________________________

Example

Part 3:

• Total subscriber capacity = 7500 × 20 = 150,000

------------------------------------------------------------------------------------------------------------------------


Or = 39.68 × 20 × 12.6 × 15 = 150,000.

• Total coverage area = 20 × 12.6 = 252 km2.

| University of Dar es Salaam | Department of Telecommunications Engineering |

_______________________________________________________________________________

Approaches To Cope With Increasing Capacity

• Adding new channels

• Frequency borrowing – Frequencies are taken from adjacent cells by congested cells

------------------------------------------------------------------------------------------------------------------------


• Microcells – Antennas move to buildings, hills, and lamp posts

• Cell Splitting – Allows an orderly growth of the cellular system.

– Increases the number of base stations in order to increase capacity.


_______________________________________________________________________________

Approaches To Cope With Increasing Capacity

• Sectoring

– Uses directional antennas to further control the interference and

frequency reuse of channels.

– Rely on base station antenna placements to improve capacity

------------------------------------------------------------------------------------------------------------------------



by reducing co-channel interference.

• Coverage zone approaches

– Distributes the coverage of a cell and extends the cell boundary

to hard-to-reach places.


by reducing co-channel interference.


_______________________________________________________________________________

Microcellular Systems

------------------------------------------------------------------------------------------------------------------------




_______________________________________________________________________________

Why Microcellular Systems

We want

We have

• Limited radio spectrum allocation

------------------------------------------------------------------------------------------------------------------------


We want

• coverage

• capacityCellular

We want

• more capacity Microcellular


_______________________________________________________________________________

Microcellular Systems Defined

• Not formally defined.

• Generally, small cells < 1 km.

• Urban area.

------------------------------------------------------------------------------------------------------------------------


• Urban area.

• Base station antennas significantly lower than surrounding roof tops.

• Coverage area largely governed by effects of surrounding buildings (urban shielding).

• Lower transmit power at base station.


_______________________________________________________________________________


The lower transmit power, combined with theeffects of urban shielding, results in a shorterreuse distance, hence cell frequencies can be

------------------------------------------------------------------------------------------------------------------------


reuse distance, hence cell frequencies can bereused more often in a given area. A high densityof reuse cells will naturally increase systemcapacity.


_______________________________________________________________________________

Microcellular Systems Propagation Models

• Propagation models should take into account the clutter introduced by individual buildings.

• The capacity can be deterministically evaluated by

------------------------------------------------------------------------------------------------------------------------


• The capacity can be deterministically evaluated by studying the effective coverage area of each individual cell.

• The propagation coverage for each cell site can be obtained either statistically (by fitting curves to measured results) or deterministically by methods such as ray-tracing.


_______________________________________________________________________________

Microcell Capacity Evaluation

• First: estimate average cell coverage area.

– Assume system is noise limited, use measurements or

models to determine how the average power varies with

------------------------------------------------------------------------------------------------------------------------


models to determine how the average power varies with

distance.

– Average cell radius = average distance from the basestation

where the required signal power and outage probabilities can

be maintained.


_______________________________________________________________________________

Microcell Capacity Evaluation

• For a cochannel limited system, it is important to design the network to have the minimum practical distance between frequencies.

• This minimum distance depends on many factors, e.g.

------------------------------------------------------------------------------------------------------------------------


• This minimum distance depends on many factors, e.g.

• the number of cochannel cells in the vicinity of the centre cell,

• the type of geographic terrain,

• the antenna height,

• the antenna pattern,

• the sectorisation and

• the transmit power at each base site.


_______________________________________________________________________________

Determination of Average Reuse Distance

------------------------------------------------------------------------------------------------------------------------



_______________________________________________________________________________

Determination of Average Reuse Distance

• A wanted cell receives interference from a cochannel cell.

• With reference to the required signal to interference ratio,six cochannel cells can be assumed and then movedcloser towards the centre cell as shown in the previousfigure.

• The diagram shows the received signal level relative to

------------------------------------------------------------------------------------------------------------------------


• The diagram shows the received signal level relative tothe received cochannel interference level for a givencochannel separation distance.

• As the cochannel cells are moved closer, there will comea point where the interference ratio and outageprobabilities are no longer met. This point represents theminimum practical reuse distance.

• Based on knowledge of the cell size, R, and the reusedistance, D, the cluster size can be determined.


_______________________________________________________________________________

Case Study – Microcellular study implemented

Microcellular study

based on the

building database

for a typical UK

urban environment

------------------------------------------------------------------------------------------------------------------------


On the left is a map

of the microcellular

area - 500 × 500 m.


_______________________________________________________________________________


Antenna height = 10 m, vertical dipole is assumed.

To study coverage, transmitter placed at centre (x=225m, y=305m).

Using suitable propagation

------------------------------------------------------------------------------------------------------------------------


Using suitable propagation model, coverage for surrounding area is found as shown to the left.

o = base stationwhite = good signal levelblack = poor signal level

Signal level better in one direction.


_______________________________________________________________________________


------------------------------------------------------------------------------------------------------------------------


From the previous figure, the average signal propagation

pathloss is derived as a function of distance.

Average Street Propagation Pathloss (f = 1800MHz)


_______________________________________________________________________________


When minimum received signal level = -70dBm (10% outage),

evaluated average cell radius = 160m

average reuse distance = 600m (for certain C/I threshold)

------------------------------------------------------------------------------------------------------------------------


average reuse distance = 600m (for certain C/I threshold)

mobile receiver height = 1.5m

mobile antenna pattern = omni

cluster size = 4.68 (7)

The image cannot be displayed. Your computer may not have enough memory to open the image, or the image may have been corrupted. Restart your computer, and then open the file again. If the red x still appears, you may have to delete the image and then insert it again.


_______________________________________________________________________________

Example

A digital microcellular system has a total bandwidth of 15 MHz and operates with a re-use factor of 7 and a cell radius of 160m. The network comprises 20 basestations and each 200 kHz carrier is made up of 8 TDMA voice channels. Assume that each user represents a traffic load of 0.02 Erlangs and

------------------------------------------------------------------------------------------------------------------------


that each user represents a traffic load of 0.02 Erlangs and that the average reuse distance is 600m.

Calculate the total call and subscriber capacity in each cell and for the entire network.


_______________________________________________________________________________

Soln:

• Calls per cell = ( 15 / ( 7 × 0.2 ) ) × 8 = 85.71

• Subscribers per cell = 4,285.7

• Assuming 200kHz call bandwidth, 15MHz supports 600 calls.

• Channels are used over cluster area of 7× 0.0804 km2 = 0.563 km2

------------------------------------------------------------------------------------------------------------------------


• Channels are used over cluster area of 7× 0.0804 km2 = 0.563 km2

• Thus, total capacity = 600 / (15 × 0.563) = 71.04 calls / MHz / km2

• Subscriber capacity = (71.04 / 0.02) = 3,552 subscribers/MHz/km2

• Total subscriber capacity = 4,285.7 × 20 = 85,714

• Or = 3,552 × 20 × 0.0804 × 15 = 85,714

• Total coverage area = 20 × 0.0256 = 1.608 km2


_______________________________________________________________________________

• Example illustrates how future microcellular systems will achieve

very high traffic densities over very small coverage areas.

– System in example supports > 85,000 subscribers in 1.6 km2 area (system has capacity > 3,500 subscribers/MHz/km2).

– In contrast, the traditional cellular systems example achieved a capacity of 39.68 subscribers/Mhz/km2. However, the system offered a larger

------------------------------------------------------------------------------------------------------------------------


of 39.68 subscribers/Mhz/km . However, the system offered a larger coverage area (252km2).

– In practice, the choice of cell size and capacity must be accurately matched to the number of users in a given area.

– Future mobile communication systems are likely to employ a mixture of small cells in the cities and large cells for more rural areas. Large cells are also more appropriate for vehicular use to reduce the number of handovers expected.

11. Cellular Planning and Capacity

Documents

Transcript of 11. Cellular Planning and Capacity