Radio
Network Planning is used to identify the geographic locations of the Node B. It
is also used to identify the antenna configurations in terms of antenna type,
height, azimuth and tilt. The high-level interaction of radio network planning
with dimensioning, site acquisition, site design and site build is illustrated
below;
The
radio network planning process is preceded by system dimensioning. Dimensioning
results are used to generate an estimate of the expected site density. This
provides a guide to the number of sites which should be necessary to achieve
the coverage requirements across a specific geographic area.
If an
operator has already deployed a 2G network then it is likely to be beneficial
to re-use as many of the existing site locations as possible. Re-use of
existing sites introduces the option of sharing antenna subsystems between 2G
and 3G, i.e. feeder cables and antennas can be shared. In general, this requires
swapping existing 2G single band antennas for new 2G/3G dual band antennas (or
swapping existing 2G dual band antennas for new 2G/3G triple band antennas).
The benefit of sharing the antenna subsystem is a reduced requirement for
antennas and feeder cables. This can be important at sites which have limited
physical space or planning restrictions. Sharing feeder cables requires a
diplexor to combine the downlink signals and to separate the uplink signals. A
potential drawback associated with sharing the same antenna subsystem is that
it may limit the radio network planner’s ability to apply independent 2G and 3G
RF optimisation, e.g. if the 3G system requires mechanical antenna down tilt
then the impact upon the 2G system must also be evaluated. However, some
antennas allow independent adjustment of the 2G and 3G electrical down tilts.
There
are two fundamental approaches to 3G radio network planning: the path loss
based approach and the 3G simulation based approach.
The
3G simulation based approach to radio network planning requires a 3G radio
network planning tool. The majority of 3G radio network planning tools make use
of Monte Carlo simulations. Monte Carlo simulations are static rather than
dynamic. This means that system performance is evaluated by considering many
independent instants (snap shots) in time. A dynamic simulation evaluates performance
by considering a series of consecutive instants in time. In general, dynamic
simulations are more time consuming than static simulations. The general principle
of a static simulation is illustrated below.
Each
simulation snap shot is started by distributing a population of UE across the
simulation area. This distribution could be based upon a uniform random
distribution or a weighted random distribution. Weightings can be based upon
Erlang maps generated from the traffic belonging to an existing 2G network.
Alternatively, weightings could be environment type dependent so for example,
urban areas could be specified to have a higher traffic density than suburban
areas. Once the population of UE have been distributed, the set of uplink and
downlink transmit powers are converged. Transmit power convergence requires an
iterative process because the transmit power of one connection has an impact
upon the transmit power of other connections. For the example of two UE, the
uplink transmit power of the first UE is calculated based upon the uplink C/I
requirement and the uplink interference floor. The transmit power of the second
UE is then calculated based upon the uplink C/I requirement and the uplink
interference floor which has now been increased by the first UE. The transmit
power of the first UE then has to be recalculated because the uplink
interference floor has been increased by the second UE. Repeating these
calculations results in convergence of both the UE transmit powers and the
uplink interference floor. A similar process can be completed in the downlink
direction. The uplink and downlink transmit powers can be converged
independently although there is some dependence in the case of a UE failing to
maintain its connection. For example, if a UE has insufficient uplink transmit
power to maintain its connection then the downlink transmit power for that UE
can be cleared and made available to the remaining UE. UE which are not able to
achieve their C/I requirements are categorised as being in outage. Outage may
also be caused by factors such as inadequate Node B baseband processing
resources or reaching the maximum allowed increase in uplink interference. The
results are recorded at the end of a simulation snap shot and the process is
repeated. This allows the simulation to generate probability distributions and
to quantify the probability of certain events occurring, e.g. the probability
that a UE will be able to establish a connection at a specific location. The
number of snap shots necessary to generate statistically stable simulation
results tends to depend upon the quantity of traffic distributed during each
snap shot. Distributing relatively little traffic tends to increase the number
of snap shots required.
The
inputs to the planning tool for the 3G simulation based approach are
illustrated beside.
The
first three inputs are similar to those required by the path loss based
approach. The 3G site data may be more complex in terms of requiring greater
information to describe the Node capability, e.g. baseband processing
capability. Propagation modelling is also more complex for a 3G simulation if
slow fading is modelled. The path loss based approach makes use of a link
budget threshold which includes a slow fade margin. 3G simulations typically
model slow fading explicitly making it necessary to specify a standard
deviation and correlation factor. The correlation factor is used to specify the
coherence of the fading experienced by the signals between a UE and the set of
surrounding Node B. A high correlation factor means that when one signal is
experiencing a fade there is a high probability that the other signals will also
be experiencing a fade. The 3G simulation tool may allow correlation factors to
be configured separately for signals originating from the same Node B and
signals originating from different Node B, i.e. the correlation is likely to be
greater for signals originating from the same Node B because the propagation
paths will be relatively similar. The slow fading correlation factor has an
impact upon the soft handover gain which is maximised when soft handover radio
links are uncorrelated. Uncorrelated radio links increase the probability of
the receiver always having at least one signal which is not experiencing a
fade.
Typical
3G parameter assumptions include maximum uplink and downlink transmit powers,
common channel transmit powers, uplink and downlink noise figures, uplink and
downlink Eb/No requirements, maximum allowed increase in uplink interference,
orthogonality, soft handover window and soft handover gains. The soft handover
gain used for 3G simulations should be interpreted differently from the soft handover
gain used for link budgets. Link budgets use a soft handover gain which
includes the diversity gain resulting from both fast and slow fading. 3G
simulations usually model slow fading explicitly and so it is not necessary to
include the slow fading diversity gain within the soft handover gain parameter.
The soft handover window used for static 3G simulations tends to represent an
average of the addition and deletion windows. The addition and deletion windows
provide hysteresis in the live network to help avoid ping-pong in terms of
cells being added and deleted from the active set. It is not possible to model
this hysteresis using a static simulation tool because only snap shots in time
are considered. When a simulated neighbouring cell is between the addition and
deletion windows there is no historical information to indicate whether the
cell is already in the active set and the signal quality is decreasing or the
cell is outside the active set and the signal quality is increasing.
3G
traffic profiles are specified in terms of the services used by the population
of UE. The full range of services can be categorized as speech, circuit
switched data and packet switched data. One or more bit rates can be associated
with each of these categories. Defining accurate traffic profiles can be
difficult and it is reasonable to start a simulation exercise by modeling one
service at a time, e.g. generate coverage and capacity results for the 12.2
kbps speech service and then generate similar results for the 384 kbps packet
switched data service. This approach provides an indication of the variance
which can be expected when changing the traffic profile. Identifying the
network capacity can be an iterative process, requiring the network planner to
increase the quantity of traffic loading the network until the probability of
blocking reaches a realistic maximum. The absolute maximum network capacity can
be quantified by distributing very large quantities of traffic, but the
blocking probability will become unrealistically high. The 3G traffic profile
also requires the geographic distribution of the UE to be defined. This
includes specifying the percentage of UE which are indoors and experience a
building penetration loss.
The
3G simulation approach to radio network planning is more time consuming than
the path loss based approach. Results take longer to generate and longer to
interpret. Simulation time depends upon the size and resolution of the
geographic area, the site density and the quantity of traffic loading the
network. It is important to ensure that sufficient simulation snap shots have
been completed to generate statistically stable results. 3G simulation tools
may include functionality for a passive scan terminal. Passive scan terminals
are used to increase the rate at which graphical geographic results are
generated. At the end of each simulation snap shot a passive scan terminal is
placed within each pixel across the simulated area and connection establishment
is attempted without modifying the already converged uplink and downlink
transmit powers. This approach allows every simulation snap shot to generate a
result for every pixel.
Otherwise,
it is only possible to generate results at the pixels where UE have been
distributed. The results generated by a passive scan terminal are used to
update the graphical geographic results but typically are not used to update
the numeric results. This means that passive scan terminals can increase the
rate at which graphical results are generated, but decrease the rate at which
numerical results are generated.
The
main benefit of completing 3G simulations is the relatively large quantity of
information which is generated. This information can help guide planning
decisions as well as provide more extensive expectations of network
performance. The main results generated by a 3G simulation are presented below;
3G
simulations can also be used for investigative studies aimed at evaluating the
impact of specific network configurations. The impact of increasing sectorisation
can be quantified for a range of different antenna gains and beamwidths. Some
of the results can be used as an input to dimensioning exercises, e.g. the
inter-cell interference ratio and soft handover overheads. Studies can also be
completed to help quantify the benefit of deploying an additional RF carrier,
or the impact of using mechanical rather than electrical antenna down tilts.
