Objective 1: To define underlying factors and constraints for broadband network expansion.
Southern Ontario is used as an example to represent the variance in the factors. In a Canada-wide scope the scale is too small to properly represent the variance in the data. The project was carried out on a Canada-wide scale.
Factor 1 -- Population and Density
The first factor that must be identified is the core population count and the population density of each subdivision. The areas of Canada that struggle with access to broadband internet are also the ones with low population density (The Standing Committee on Industry, Science, and Technology, 2018). Additionally, areas with low core population counts, such as distant rural communities, have less or no access to broadband networks at all (Rob, M. 2011). This is due to the fact that broadband companies prefer to improve current broadband lines over creating new ones. Therefore, to determine where to place a new broadband distribution centre it is necessary to gather population statistics for each Canadian subdivision. More emphasis will be placed on regions that have larger population counts in a denser area. For this section, GDP will not be considered a factor as the installation of new broadband is sanctioned by the government, but the form of broadband installed is decided by the telecommunication companies. Therefore, the inclusion of GDP in this factor is mute because in many northern areas with low population and density the implementation of, for example, copper wire, would be far more difficult than necessary when a new cellular tower could sustain the low population.
Figure 2: Population density factor in Southern Ontario
Factor 2 -- Quality of Broadband Connection
The second factor that must be identified is the quality of broadband connection in subdivisions for each applicable style of broadband. The better the quality of the connection, the less likely that an upgrade is necessary. Since this paper aims to find locations that would benefit from broadband infrastructure improvements, and the Canadian government recommends that the average Canadian household has access to 50↓:10↑ (50 megabits download and 10 megabits upload) megabit connections, subdivisions with less access to broadband will be valued higher (increasing weight as distance to nearest cell tower increases), as they have more potential upgradability than subdivisions with greater access to different broadband types (The Standing Committee on Industry, Science, and Technology, 2018).
Figure 3: Cellular tower distance from cell tower nodes across Canada
Factor 3 -- Distance from urban centre
The third factor that must be identified is the proximity of subdivisions to large urban centres. Canadian telecom companies are much more likely to improve urban areas than they are to develop rural areas (Canadian Radio-television and Telecommunication Commision, 2018). Urban centres can be considered the backbone of Canada’s broadband network (The Standing Committee on Industry, Science, and Technology, 2018). subdivisions which are closer to urban centres can have different installation/maintenance costs (EUCNCT, 2015).
Figure 4: Distance buffer from all urban centres across Canada
Factor 4 -- Existence of broadband infrastructure
The fourth factor that must be identified is the existence of broadband infrastructure. It is important to connect as many communities to the internet as possible, due to the social and economic benefits it brings. Regions which have no access to the internet will have greater benefit from a line installation than regions which already have access, and as such will be weighted appropriately, according to a pairwise weighting scheme outlined in figures 2 through 4. This factor, as well as, factor two use distance to attribute different weights based on distance. The purpose is to give greater weight to areas which both exist further from the nearest broadband tower, while factor two gives greater weight to areas of distance from urban centers. The culmination of these two layers is the prioritization of areas within 100km of urban centers which suffer from the poorest connection.
This data is best modelled in GIS as distance buffers, which use specified average signal range to create a layout for all three types of broadband. Specifically, 50km range buffers for cellular towers, 5km buffers for copper wire transmission and Fibre Optics would be limited to 25km. A cellular range of 50km is chosen because, according to the 3GPP TS 05.10 for GSM cellular towers, the maximum range for a tower is approximately 35km (3GPP, 2003). However, this value can be expanded to a range of approximately 120km for remote locations according to 3GPP TS 45.010 (3GPP, 2010). As such, a 50km tower range is deemed appropriate, because it closely matches the original tower range value, while accounting for longer ranges needed for long distance data coverage.
Figure 5: Cellular connectivity regions in Southern Ontario
Figure 6: Copper connectivity regions in Southern Ontario
Figure 7: Fibre connectivity regions in Southern Ontario
Factor 5 -- Upgradability of broadband infrastructure
The fifth factor is the upgradability of internet infrastructure within the subdivision. Broadband internet speeds differ greatly between mediums, with wireless speeds being the slowest, followed by copper wire, then fibre wire. (EUCNCT, 2015). When upgrading, the assumption is that the upgrade will happen linearly, such that an upgrade from wireless will go to copper, and copper will go to fibre. This means that the weighting for this factor will favour wireless upgrades over copper upgrades.
Figure 8: Regions with all forms of broadband connection in Southern Ontario