Multilayer, locality aware, telecommunication network deployment algorithm

Abstract

Purpose – In this paper we propose an iterative approach for the deployment of rural telecommunication networks.
 Methodology/approach/design – This approach relies heavily on the concept of locality, prioritizing small ‘cells’ with a considerable population density, and exploits the natural nesting of the distribution of rural communities, focusing in communities which are populous enough to justify the investment required to provide them with connectivity, and whose sheer size promotes the formation of ‘satellite’ communities that could be benefited from the initial investment at a marginal expense. For this approach, the concept of ‘cells’ is paramount, which are constructed iteratively based on the contour of a Voronoi tessellation centered on the community of interest. Once the focal community has been ‘connected’ with network of the previous layer, the process is repeated with less populous communities at each stage until a coverage threshold has been reached. One of the main contributions of this methodology is that it makes every calculation based on ‘street distance’ instead of Euclidean, giving a more realistic approximate of the length of the network and hence the amount of the investment. To test our results, we ran our experiments on two segregated communities in one of the most complicated terrains, due to the mountain chains, in the state of Chiapas, Mexico.
 Findings – The results suggest that the use of ‘street distance’ and a local approach leads to the deployment of a remarkably different network than the standard methodology would imply.
 Practical implications – The results of this paper might lead to a significant reduction in the costs associated with these kinds of projects and therefore make the democratization of connectivity a reality. In order to make our results reproducible, we make all our code open and publicly available on GitHub.