Source Code Description for Paper “Privacy-friendly Decentralized data Aggregation for Crowdsensing Networks”

This is a brief description for the source code of the paper “Privacy-friendly Decentralized data Aggregation for Crowdsensing Networks” which is included in IEEE globecom 2020.

File Description

There are five executable python files in this repo, namely

• draw4networks.py,
• draw_privacy_acc_var.py,
• draw_histo.py,
• draw_netsz_itt_acc.py, and
• draw_real_word_test.py,

which seperatively corresponds to:

• generating random geometric graphs,
• depicting privacy-accuracy trade-off for the proposed algorithm and calculating variance,
• plotting histograms of empirical convergence value,
• showing accuracy, the number of iterations to reach a convergence as a function of network size, and
• drawing convergence process of each terminal,

with other files being the implementation of the proposed algorithm, the original classical algorithm and some basic functions.

Requires

• Both Windows or Linux is ok (operating system-independent).
• At least python 3.x is required (please also ensure the pakage networkx is installed).

How to run

With the above reqiures met, just enter the command python <file_name>(replacing <file_name> with your target file name) for running the programms.