Driving telecommunications business changes with GIS
Network Design / Engineering
Broadband networks are complex arrangements of digital terminals, optical, coax, and copper signal transmission
media, power cables, optical/electrical signal converters, amplifiers, taps, remote terminals, transformers and power
inserters. The configuration of these elements is subject to varied design disciplines and algorithms. Serving areas
are well defined geographical regions served by their owndistributioninfrastructure. The initialtask entailsthe
definitionof these areas. Welldefinedserving areas will offer effective subscription bases measured by an
appropriate balance between anticipated utilization, cost effectiveness, and regional consistency. In addition to
economic and engineering factors, serving areas may be constrained by town and other political boundaries.
Definition of serving areas is a task uniquely suited to GIS.
The placement of equipment within these regions determines the overall behavior of the resulting network, because
virtually every aspect of the equipment performance is governed or constrained by the physical distances over
which the signals must be transported.
The ideal solution to the network design problem is the combination of powerfi.d engineering design and analytical
tools and Geographical Information System technology. The former must be inherently flexible in order to
accommodate the rapid evolution of the broadband technology. The complexity of the network, and the need to
fi-equently accommodate new types of equipment, means that an object-oriented approach has significant
advantages. The development of effective designs is a difficult problem, and experience suggests that an
appropriate combination of automation and human intervention provides the most effective solution in terms of both
design quality and productivity. The design process consists of the following five major steps:
- Serving area assessment and geographical definition.
- RF design of the distribution (fiber node) area.
- Powering of the distribution equipment including Radio Frequency (RF) adjustments if necessary.
- Fiber serving arrangements.
- Central Office power feeder design.
The first phase of the implementation at SNET provides practical support for step 1 and comprehensive solutions to
steps 2 and 3. The second phase of the implementation, currently underway, will address steps 4 and 5.
Engineering Design Capabilities
The SNET broadband engineering design tool (BEDT) was developed by Smallworld Systems, Inc., together with
their business partner Geodata Solutions, based on specifications developed by Bellcore and SNET. Key features of
the design tool include:
- An Oracle based Corporate equipment catalog which stores equipment information ranging from basic
attributes such as model name and manufacturer, to accounting information such as labor class and
account code, to detailed engineering information such as insertion loss at different frequencies.
- Automated network layout using comprehensive strand map information developed from a
combination of photogrammetry and field survey.
Comprehensive Amplifier modeling. Amplifiers are devices which boost the level and adjust the
frequency response of the RF signal supplied to them. Standard engineering practice establishes their
output signal to a predefine level which may vary depending on the type of amplifier and its position
in the amplifier cascade. This predefine output level needs to be achieved regardless of the input
signal, which may vary significantly. The output signal is adjusted to the required level by the
insertion of plug-in components which modify the signal in different ways. Typical plug-in
- components include pads, equalizers, splitters, cable simulators and flatness boards. In order to
accommodate the latest generation of amplifier technology, a powerful and flexible approach to
handling amplifiers was developed. With this approach, the internal structure and internal connectivity
of the amplifier within the engineering design tool using Smallworld’s concept of multiple worlds.
When an amplifier is created, its internal representation is copied from a template for that amplifier
model. The appropriate plug-in components are then automatically determined by tracing through the
internal representation from the input port to each output port on the forward path; and vice versa on
the reverse path. Plug-in selection from the equipment catalog ensures that the available input signal is
transformed into the required output signal.
- Power Validation. BEDT includes integrated powering validation tools, which run iterative
calculations to determine voltage and current at each active device on the network and checks if the
values fall within the acceptable range for each device. The system displays graphical feedback on
portions of the network which can be powered successfully and those which cannot using color. It
also displays a list of all active devices with their voltage and current. This list can be used to navigate
directly to any device.
- Multi-page Plotting. A bill of materials and a set of work prints is provided to the construction crew
for each completed design. In general, the area of a typical design is too large to fit legibly on a single
map sheet, so BEDT includes functionality to automatically generate multi-page plots. A single “plot
design” button determines the extent of the design and automatically subdivides it into an appropriate
number of map sheets at a user specified scale. Itthengenerates an overview or index plot showing
the whole area, together with detailed plots of the map sheets which have been defined. The detailed
map plots each contain references back to the overview plot (such as “Sheet 7 of 16) and also show the
numbers of the adjoining detailed plots. A typical example of an overview plot is illustrated below:
The overview map includesa summaryof the materialsrequiredto constructthe job. A detailedbill of
material is also generated. It can be sorted and formatted based on financial and administrative criteria
such as labor classes, construction type, and account codes; and supplies the necessary input to the resource
and inventory management processes. All of this information is drawn directly ffom the corporate Oracle-based
equipment catalog.
Managing Address Information
A network design is useless unless the relationship between the network and the customers it serves can be
accurately established. The purification of service address information and its association with physical buildings
is, therefore, critical. So how do we go about the process of establishing these relationships and what role does a
GIS play? Consider the overview provided in Figure 2. Existing service address records are pre-scrubbed and
loaded into a portable unit in conjunction with the corresponding section of the GIS database. Field operations must
provide:
- Information about missing and new support structures, such as poles and pedestals.
- Associations between service addresses and structures, typically buildings, on the GIS base.
- Updates and corrections when address anomalies are detected.
If final corrections to this information were made prior to its import into the GIS base then we would miss a
significant opportunity to improve the assessment and administration of address information. This can and should
be conducted against a detailed GIS backdrop of towns, streets and so on. This allows us to view and manage
address information within a powerfil geographical context and thereby maximize its accuracy and consistency.
Daily operations require that field personnel navigate to service addresses and equipment sites to perform service
activations, routine maintenance and to perform necessary repairs. Equipment sites are locations which support
equipment but have no underlying postal address. An equipment bearing pole is perhaps the most obvious example.
In order to eliminate address duplication and to optimize the management of address changes, equipment sites will
be logically linked to the nearest address bearing entity, typically a building or a street, and derive their address
attributes accordingly. This will enable the set of address changes at any point in time to be readily determined and
propagated to down stream systems,
Figure 2: Initial and ongoing address grooming and administration
