Chapter 7: Infrastructure Feasibility Algorithms
7.1: The Need for Algorithms and the Definition of Feasibility
The purpose of this thesis is not only to provide a guide to broadband Internet infrastructure technologies, but also to provide some way of assessing the feasibility of these technologies in specific regions and communities. Such a resource would be a valuable tool for municipalities, community groups and individuals, and can help them to discover and differentiate between the available technologies.
There are two major types of broadband Internet infrastructure that have been covered by this thesis and are relevant for discussion at the community level: pipelines and last-mile networks. Last-mile networks are the part of the infrastructure that connect individual homes or businesses to a central medium (e.g. Canopy cluster, phone lines & DSLAM, cable lines & CMTS, etc.). Pipelines are high-capacity links which connect a last-mile network's central medium to a major regional backbone, either directly or possibly relayed by a satellite. For the purposes of this thesis, “backbone” refers to a major fibre network, often spanning the country, which several communities are connected to. “Pipeline” or “backhaul” refers to a community's high-capacity link to a backbone. Another unique term used in these algorithms is “target area”. This refers to the subset of a community or region that the user of the algorithm needs to find a feasible pipeline and last-mile network for. Pipelines and last-mile networks are physically independent of each other; that is, any pipeline technology can be used with any last-mile network technology. For this reason, this thesis will present two separate feasibility algorithms: one to determine the feasibility of pipeline infrastructure, and one to determine the feasibility of last-mile network infrastructure.
Additionally, it is important to define feasibility. From the perspective of a community, there are at least two relevant types of feasibility: physical feasibility and economic feasibility. Physical feasibility, for the purposes of this thesis, is the quality of being accomplishable according to the laws and rules of physics and engineering. If a project were physically unfeasible, something like the inability for wireless signals to travel through trees or the inability for fibre to be installed across a rocky mountain might be preventing it from being physically feasible. Economic feasibility, for the purposes of this thesis, is the quality of being accomplishable according to the laws of economics. If a project were economically unfeasible, something like having a project cost of $1,000,000 and a budget of $50,000 might be preventing it from being economically feasible. Physical feasibility and economic feasibility are often independent of each other and need to be computed separately.
The two algorithms presented in this thesis only deal with physical feasibility; the determination of economic feasibility is left for the user of the algorithm to determine. This approach was decided upon for reasons of simplicity and accuracy. Broadband technologies each have basic physical requirements that can be defined as sets of rules. If any of the rules are violated, the technology becomes physically unfeasible. Examples of these rules may include the requirement of the preexistence of some kind of infrastructure, or the requirement of a clear line of sight between two points. Thus, individual technologies can be eliminated from the list of feasible technologies using simple, first-order logic. There are few, if any, variables that come into play here. By contrast, economic feasibility is highly subjective and highly variable. The cost of broadband infrastructure equipment is always changing, and surprisingly difficult to get recent data on. Much of the pricing stated in this thesis is only estimated, or has even been guesstimated. One probable indicator of economic feasibility would be the cost per user for the deployment of infrastructure. Uniquenesses in a community's geography, existing infrastructure, or demographics might make such a calculation highly inaccurate. One number may be economically feasible for one community, while completely infeasible for another. The fluctuating and possibly inaccurate pricing for these technologies adds to this inaccuracy. Because no reasonable level of accuracy can be guaranteed by a generalized economic feasibility algorithm, no such attempt has been made by this thesis. Instead, the physical feasibility algorithms will calculate an estimated minimum and/or maximum deployment cost for each technology and present it to the user. It is left up to the user to determine if these numbers may be economically feasible for their unique situation.
© Jake Cormier, 2006 [jake (at) stormcloudstudios.com]
Completed as a partial requirement for the degree of Bachelor of Science (specialized)
Department of Computer Science :: Algoma University College :: Sault Ste. Marie, Ontario :: Spring 2006