New Technique for Proposing Networks Topology using GPS and GIS

The problem of proposed topology for network comes when using Prim’s algorithm with default distance (unrealistic distances) between network’s nodes and don’t care about the lakes, high hills, buildings, etc. This problem will cause incorrect estimations for cost (budget) of requirements like the media (optic fibre) and the number or type of Access-points, regenerator, Optic Amplifier, etc. This paper proposed a new technique of implementing Prim’s algorithm to obtain realistic topology using realistic distances between network’s nodes via Global Positioning System GPS and Geographic Information Systems GIS packages. Applying the new technique on academic institutes network of Erbil city from view of media (optic fibre) shows that there is disability in cost (budget) of the media which is needed (nearly) 4 times if implement default Prim’s algorithm (don’t using GPS & GIS) base on unrealistic distances between the nodes.

💡 Research Summary

The paper addresses a critical shortcoming in conventional network topology design that relies on the classic Prim’s algorithm for constructing a Minimum Spanning Tree (MST). While Prim’s algorithm guarantees the least‑cost interconnection when supplied with accurate edge weights, most practical deployments substitute those weights with straight‑line or approximate distances that ignore real‑world obstacles such as lakes, hills, and built‑up areas. This simplification leads to severe underestimation of material costs (e.g., optical fiber length), the number and placement of repeaters, optical amplifiers, and other ancillary equipment, ultimately causing budget overruns and project delays.

To remedy this, the authors propose a two‑stage methodology that integrates Global Positioning System (GPS) measurements with Geographic Information System (GIS) analysis. In the first stage, precise latitude‑longitude coordinates of each network node (in the case study, ten academic institutions in Erbil, Iraq) are collected using handheld GPS receivers, achieving positional accuracy within a few meters. In the second stage, these coordinates are imported into a GIS platform (ArcGIS) where they are overlaid on high‑resolution spatial layers that include elevation models, hydrography, road networks, and land‑use classifications. Using the GIS Network Analyst toolbox, the shortest feasible path along existing roadways and pedestrian routes is computed for every pair of nodes, automatically accounting for elevation changes and required detours around natural or built obstacles.

The resulting “real‑world distances” replace the naïve straight‑line values as edge weights in Prim’s algorithm. The algorithm itself remains unchanged; only the input matrix is updated. When applied to the Erbil academic network, the authors observed that the total fiber length required by the realistic MST is roughly four times greater than that estimated using default Euclidean distances (approximately 4,800 m versus 1,200 m). Consequently, the projected cost for fiber procurement, trenching, and installation, as well as the number of required optical repeaters and amplifiers, escalates by a comparable factor. The study also highlights that certain high‑elevation or water‑crossing segments necessitate additional equipment, further inflating both capital and operational expenditures.

Key insights derived from the work include: (1) Incorporating GPS‑derived coordinates and GIS‑based path analysis dramatically improves the fidelity of cost estimates, turning speculative budgeting into data‑driven forecasting. (2) The integration can be achieved with relatively low‑cost hardware and existing GIS software, making it accessible to municipal planners, telecom operators, and academic institutions. (3) Because the underlying MST algorithm does not require modification, the approach can be retrofitted into existing design workflows with minimal disruption. (4) The methodology is not limited to fiber‑optic networks; it can be generalized to power distribution, water supply, transportation, and any infrastructure where physical routing constraints dominate.

The authors conclude by outlining future research directions: developing automated pipelines for large‑scale citywide GPS data acquisition, incorporating real‑time traffic and construction updates to enable dynamic re‑optimization of the MST, and extending the model to multi‑objective optimization that balances cost, reliability, and latency. Such advances would support the broader vision of smart‑city infrastructure planning, where accurate spatial analytics are integral from the conceptual design phase through to on‑the‑ground deployment.