SPARTA🗡️-GEMSTONE💎: A two-phase approach for efficient node placement in 3D wireless sensor networks under $Q$-Coverage and $Q$-Connectivity constraints

This repository is the official implementation of the graduation thesis:

Truong Vu, SPARTA-GEMSTONE: A two-phase approach for efficient node placement in 3D wireless sensor networks under Q-Coverage and Q-Connectivity constraints.

[Thesis][Slides]

Result of SPARTA-CC-GEMSTONE with the default setting.

1. Clone the repository

Now please clone the repository and switch to the root directory:

git clone https://github.com/truong2710-cyber/SPARTA-GEMSTONE.git /your/path/sparta_gemstone
cd /your/path/sparta_gemstone

2. Environment setup

We recommend start from a fresh Python environment and install the required packages to avoid incompatibility issues. We provide an environment config for you in environment.yml. You can create the environtment and activate it by:

conda env create -f environment.yml
conda activate spargem

3. Run experiments

We provide some arguments for the experiments as follows:

• --N: number of targets, default = $400$.
• --qm: maximum value of $Q$ for every target, default = $10$.
• --rs: sensing range, default = $40 (m)$.
• --rc: communication range, default = $80 (m)$.
• --d: dataset name, which is in the list ["bacgiang", "hanoi", "lamdong", "sonla", "thaibinh"], default = sonla.
• --w: the width (also the length) of the AoI, default = $2000(m)$.
• --b: base station location in xy-coordinate, separated by a single comma, default = 0,0.
• --ch: which factor to change. There are four options ['n', 'Rs', 'Q', 'Rc'], default = n.
• --s: the step of the change, default = $15$.
• --ns: the number of steps to take, default = $6$.
• --nr: the number of runs in one data config, default = $20$.
• --p1: the algorithm to use in phase I, default = sparta_cc. There are three options ['sparta', 'sparta_cc', 'sparta_cp'].
• --p2: the algorithm to use in phase II, default = gemstone. There are two options ['gemstone', 'cmfa'].

Run file main.py to execute the experiment. For example:

python main.py --N 400 --qm 10 --rs 40 --rc 80 --d sonla --w 2000 --b 0,0 --ch n --s 15 --ns 6 --nr 20 --p1 sparta_cp --p2 gemstone

The result is saved in the folder Result/dataset_name/algo_name in a .txt file.