This repository contains all code, scripts and datasets for recreating the evaluating cluster based MEC server placement algorithm

The work has been published in MECOMM 2018. You can cite our work as follows:

Nitinder Mohan, Aleksandr Zavodovski, Pengyuan Zhou, and Jussi Kangasharju. 2018. Anveshak: Placing Edge Servers In The Wild. In Proceedings of the 2018 Workshop on Mobile Edge Communications (MECOMM'18). ACM, New York, NY, USA, 7-12. DOI: https://doi.org/10.1145/3229556.3229560

Requirements

The code requires you to have the following packages installed.

1. Python v3
2. GraphLab
3. R

Directory Structure

• Dataset: Contains required raw basestation and WiFi AP density, user requests and GPS coordinate data files.
• Edge_Server_Clustering: R-based code files which creates density clusters of users and maps its to a physical location in a polygon og GPS coordinates.
• MilanoGridMap.ipynb: Python code which maps WiFi Access points dataset to Milano grid (area of study in this work).
• Map Convert.ipynb: Tags WiGle dataset of WiFi APs to GPS coordinates

The code is commented and is self-explanatory. Still if you have any issues, feel free to reach me at nitinder.mohan@helsinki.fi