Fear of Interference: Dynamic Channel Allocation

Fear of Interference
By Andrew Miceli
Techniques to increase capacity also may increase co-channel interference.
Although interference problems in wireless networks can come from outside

sources, most engineers first must focus on the interference their own
network design creates, co-channel interference.
In AMPS, GSM and IS-136 networks, calls are assigned specific frequencies. In
GSM and IS-136, calls also are assigned specific timeslots within those
frequencies. Design engineers are careful to plan which frequencies can be
used at which base stations, the idea being to keep an adequate distance
between the reuse to ensure they won't interfere with each other. This is co-
channel interference.
Traditionally, design engineers have used the “pattern of 7” to lay out which
sites could use which frequencies. In this pattern, frequencies are reused only
two base stations away. With an overall fixed amount of available spectrum,
this reuse pattern then becomes the limiting factor for capacity. Lower the
reuse pattern, and you can increase capacity, but you also increase the
possibility of co-channel interference.
Infrastructure vendors obviously realize the importance of mitigating co-

channel interference to increase capacity. A basic feature is alternating
channel assignments. If a set of frequencies simply were assigned to specific
sectors, as requests for channels came in, the base station always would start
with the first frequency and work its way up. That first frequency almost
always would be up and would have the most use on it. If the co-channel site
had those same frequencies assigned in the same way, they both would be
transmitting on the same frequencies all the time. By alternating the
assignments — in other words, one sector assigns from the top of the list
down while the other starts at the bottom and works up — as long as all of the
channels are not used, co-channel interference can be limited.
Dynamic Channel Allocation

Most infrastructure vendors also have implemented a dynamic channel-
allocation feature in their networks. Essentially, this feature allows a set of
frequencies to be assigned to base stations and allows the network to choose
which frequency channels to use for each call across a cluster of sites.
Without this feature, a set of frequencies would be assigned to each sector,
but all might not be used simultaneously, depending on demand.
With dynamic channel allocation, the network is able to determine what

channels are free in a specific cluster of sites at the time of the call and assign
the channels dynamically as needed. This can help design engineers use co-
channels closer together in many cases. Related to this is the concept of
frequency hopping, in which, during the process of a call, multiple channels
are used to spread the transmissions among many channels, again with the
idea of negating the effects of interference on any given channel.
The first step in troubleshooting co-channel problems is to identify that you

have a co-channel problem in the first place. Such interference would
manifest itself in the form of RF losses and handoff failures, in addition to poor
call quality. In digital systems, there also will be an increase in the bit error
rate (BER).
Network design can cause two specific types of interference — co-channel, as

described above, when the same channel is used too close together, and
adjacent channel, when the adjacent channel is used too close together.
Adjacent-channel interference is essentially the same as co-channel, except
that the interferer generally must be higher in power than the signal of
interest to cause problems, and, of course, the interfering signal is in the
adjacent channel as opposed to the same channel.
There is a variety of opinions on when co-channel interferers begin to cause

problems. Generally, in IS-136 systems, most people choose 17dB below the
signal of interest as a good threshold. Drive-test gear available today can
measure the power from specific base stations at any given location. In this
way, an engineer can measure the power of both base stations and determine
whether there is a co-channel problem. The obvious solutions are changing
frequency assignments; however, antenna re-orientations, power and
parameter changes also can be used.
Pilot Pollution
In IS-95 and 3G CDMA-based wireless formats, co-channel interference is
inherent to the network design at a couple of levels. Each user is assigned a
code instead of a specific frequency. In this way, all of the users actually
transmit in the same frequency channel. A complex power-control system in
these formats helps ensure that any one user is not overpowering another
user.
CDMA networks are designed so a set amount of users still can maintain a
quality uplink and downlink in a cell's coverage, even though all of the power
for all of the users is at the same frequency. Any extra energy in the band —
for example, energy from the adjacent cells that are transmitting at the same
frequency — can cause problems. This is mitigated with the soft-handoff
concept: a mobile can receive signals from separate base stations
simultaneously, combining them to actually create a more robust connection.
Too much of anything is never a good thing. If more than four base stations
reach a mobile with enough power, the mobile will not be able to receive all of
them, and the excess energy only will add to the noise floor, thus lowering the
signal-to-noise ratio, which dictates the BER — or call quality. In CDMA circles,
this is called pilot pollution rather than co-channel interference.
Pilot pollution generally is the result of putting too many cell sites too close
together at too high a transmit power. In CDMA network design, capacity is
directly related to how close the cell sites are to each other. Like the 2G TDMA
systems discussed previously, design engineers must move cell sites closer
together to gain capacity. Again, the trade-off is the potential for pilot
pollution.
Pilot pollution can be identified with today's drive-test gear fairly easily,
specifically with the use of an independent PN scanner. This scanner can look
at the power in all of the PN offsets, the specific code offset that identifies
base stations in CDMA networks. Via post processing, these tools then can
show locations where the condition exists. Often in CDMA, the solution is to
lower pilot power, although antenna re-orientations are also common
solutions.
Uplink Issues
Pilot pollution is related specifically to the downlink in today's CDMA systems.
The uplink also has concerns that are related to co-channel use. Depending on
the cell radius, only a certain number of wireless phones will be able to
transmit in the same frequency channel before the base station will begin to
have trouble receiving the signals. Even though the complex power-control
systems in CDMA allow all of the handsets to transmit at a power level such
that at the base station they are all equal, as the base station is receiving the
signal from one phone, all of the other phones' energy adds to the noise of
the received S/N ratio.
At some point, this noise rise will be so high that the S/N ratio will not be high
enough to support receiving the call adequately. Many consider this noise rise
the capacity-limiting factor of a cell. This is one of the main reasons for the
recent proliferation of tools designed to make the handsets transmit at lower
powers, yet with high call quality.
For example, the superconductor-based filter and low-noise amplifier system

can lower the handset transmit powers by as much as 3dB to 5dB by
increasing the base station's sensitivity. In this way, more phones can be
established on a call before the noise level increases too much, and the uplink
capacity is increased.
As carriers try to squeeze more capacity out of 2G TDMA and CDMA systems,
the drive will be to deploy more sites closer together. This means co-channel
interference, whether on the uplink or the downlink, may surface.
Understanding this up-front in the network-design phase, whether in a brand
new digital overlay or simply a capacity offload deployment, is the easiest
way to head off problems later. Troubleshooting the condition, however, is not
terribly difficult with the tools available today, assuming over-burdened
performance engineers have the time available to perform such tasks.

Fear of Interference: Dynamic Channel Allocation

Uploaded by

Copyright:

Available Formats

Fear of Interference: Dynamic Channel Allocation

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fear of Interference: Dynamic Channel Allocation

Uploaded by

Copyright:

Available Formats

Fear of Interference

Although interference problems in wireless networks can come from outside

Infrastructure vendors obviously realize the importance of mitigating co-

Dynamic Channel Allocation

With dynamic channel allocation, the network is able to determine what

The first step in troubleshooting co-channel problems is to identify that you

Network design can cause two specific types of interference — co-channel, as

There is a variety of opinions on when co-channel interferers begin to cause

For example, the superconductor-based filter and low-noise amplifier system

You might also like