Section 6 Wideband CDMA Radio Network Planning

1
Section 6 Wideband CDMA Radio Network Planning
2

Radio Network Planning

• A radio network planning consists of three
  phases
• Network Dimensioning (using link budgets)
• 2. Detailed capacity and coverage planning
  (using planning tools)
• 3. Network optimisation (using optimisation
  tool)

3
Phase 1 Network Dimensioning

• Dimensioning the WCDMA radio network includes
  radio link budget and coverage analysis, capacity
  estimation and estimation of the amount of
  network equipment (such as number of BSs and
  RNCs) required.
• These estimations will be based on the operators
  requirements on coverage, capacity and quality of
  service.

4

• WCDMA-specific parameters in the link budget
  
  compared to those parameters
  used in a TDMA-based radio systems are
• -Interference margin
• The value of the interference margin used
  in the link budget depends on the loading of the
  cell. Higher is the value of the interference
  margin in the uplink, the smaller is the coverage
  area. Typical values are 1.0-3.0 dB in the
  coverage-limited cases, corresponding to 20-50
  loading.

5

• -Fast fading margin
• For slow-moving mobiles, to take care of
  fast fading effect, a fast fading margin in the
  range of 2.0-5.0 dB should be included in the
  link budget.
• -Soft handover gain
• Due to uncorrelated channels from the MS
  to the BSs, handover gives a gain against slow
  fading. Also, soft handover gives an additional
  macro diversity gain against fast fading. The
  total handover gain can be assumed to be in the
  range of 2.0-3.0 dB.

6
Link budget approach
Coverage requirement for a specific data rate
with uniform load
Derive Link Budget
Input existing 2G sites that can be Upgraded to
3G
Refine design, put new sites using Planners
individual judgment
Coverage satisfied?
No
Yes
End
7
Uplink Link Budget Example
8
(No Transcript)
9
(No Transcript)
10
Cell range From
the link budget, the cell range R can be easily
calculated using a known propagation model, for
example the Okumura-Hata model. The Okumura-Hata
propagation model for an urban macro-cell with
base station antenna height of 30m, mobile
antenna Height of 1.5m and carrier frequency of
1950 MHz is given by L 137.4
35.2 where L is the path loss in dB and R is the
cell range in Km. For suburn areas we assume an
additional area correction factor of 8 dB and
therefore the path loss is L
129.4 35.2

11

• Some Definitions
• Ratio of other cell to own cell interference
• In the uplink, it is calculated for the BS,
  therefore i is similar for all
• connections within one cell. However in the
  downlink, it is calculated
• for each MS and therefore depends on the MS
  location.
• i ranges from 0.15 (very well isolated
  microcells) to 1.2 ( poor radio
• network planning.)

12

• For the downlink, i is defined as
• i
• where is the power received from
  other BSs and pj is the power
• received from the serving BS.
• Noise rise
• noise rise

13

• Capacity estimation
• The second part of dimensining is to estimate the
  capacity per cell i.e.,
• supported traffic per BS. The capacity per cell
  depends on the amount
• of interference per cell, hence it can be
  calculated from the load equations.
• Uplink load factor equation
• 
  (1)
• 
  
• where W is the chiprate, pr,j is the received
  signal power for mobile user j,
• is the activity factor of user j, Rj is the
  bit rate of user j and the
• total received wideband power including thermal
  noise power in the BS.

14
Equation (1) can be rewritten as

(2)

we define where is
the load factor of one connection. Using this
equation and equation (2), one can obtain
as

(3)



15
The total received interference, excluding the
thermal noise ,can be written as

(4)


The noise rise is defined as
Noise rise
(5) and using
(4), we can obtain
16
Noise rise

(6) where is defined as
the uplink load factor and equals to


(7) when becomes close to 1, the
corresponding noise rise approaches to infinity
and system has reached its pole capacity. If
the interference from the other cells is taken
into account, then one can write


17




(9)

where i is the ratio of other cells to
own cell interference. The interference margin
used in the link budget must be equal to the
maximum planned noise rise i.e., -10 log(1-
). For an all voice service network, where all
N users in the cell have a low bit rate of R, we
can write
18
and hence equation (9) is simplified to
19
- Downlink load factor In the absence of
intra- and inter- cell interferences, one can
write In the absence of interferences, we
defined and hence,
20


when we take into account both intra- and
inter- cell interferences, we have where
is the orthogonality of the channel of mobile
user j. Its value depends on the channel
multipath fading where 1 means no
multipath fading. is the ratio of other cell
to own cell base station power, received by the
mobile user j.
21
The downlink load factor is defined
as since, in the uplink, i and
depends on the location of the mobile user and
they should therefore, be approximated by their
average values across the cell, and
.
22
The average value of the downlink load can then
be approximated as the noise rise is given
by noise rise
Interference margin when
1 noise rise the system
approaches its pole capacity.
23
Total BS transmission power
The total BS transmission power can be written
as where is the average attennation
between the BS and mobile receiver (6 dB less
than the maximum path loss) since
24
and then where
is the power spectral density of the
mobile receiver and is given by where F is the
noise figure of the mobile receiver with typical
values of 5-9 dB.
25
Throughput per
cell where N is the number of users per cell,
R is the bit rate and
is the block error rate.
26

• Link budget approach
• Pros
• - Enables fast planning of coverage for a
  pre-specified uniform load
• - Skilled 3G staff not a requirement
• Cons
• - Too simplistic for WCDMA where
  coverage/capacity/QoS are
• closely related
• - The final performance of the network
  cannot be derived based on
• this method
• - Mix of traffic cannot be taken into
  account

27

• Phase2 Detailed capacity and coverge planning
• In this phase, real propagation data from the
  planned area and the
• estimated user density and user traffic are
  used.
• The output of this phase are the base station
  locations, configuration and
• network parameters.

28
Static simulation approach
Coverage/traffic/QoS requirements
Input existing 2G sites which can be upgraded to
3G
Refine design, put new sites using Planners
individual judgment
WCDMA static simulator
Coverage/capacity/QoS Satisfied?
No
Yes
End.
29

• Static simulation approach
• Pros
• - Average QoS, capacity and coverage may be
  assessed for a mix
• of traffic
• Cons
• - Can only be run on a limited area, typical
  figures for running time
• for a 3 Km x 3 Km area is 5-8 hours on a
  Unix work station
• - Manual judgment must be exercised in
  interpreting the results and
• making decisions to improve the plan.
• - Plans may need to be iterated several times
  (on average 5 times)
• before the desired capacity/QoS/ coverage
  is achieved. This takes
• total planning time for a 3 Km x 3 Km to 1
  to 2 working days at best!
• - Skilled 3G a prerequisite

30

• Phase 3 Optimisation Phase
• Network optimiser
• Optimises WCDMA FDD network plan minimising the
  number of sites
  required to achieved the coverage/traffic/QoS
  targets set by the user.
• An Optimiser also automatically selects the most
  appropriate antenna tilt, direction and
  sectorisation in order to achieve the required
  coverage/traffic/QoS.

31
Network optimiser
Feed in your site portfolio
Set optimisation criteria
Run Optimiser algorithms
End
32
Optimisation phase
Coverage information
Optimised site locations
WCDMA FDD parameters
Optimiser
Traffic information
Coverage, Capacity/QOS statistics
Site locations
Optimisation criteria
33

• Reference
• WCDMA for UMTS, Edited by Harri Holma and Antti
  Toskala,
• Second edition, John Wiley Son Ltd, ISBN
  0-470-84467-1.
• 
  
•