Intrusion Detection System in Wireless Sensor Network using Machine Learning

One of the fields where Artificial Intelligence (AI) must continue to innovate is computer security. The integration of Wireless Sensor Networks (WSN) with the Internet of Things (IoT) creates ecosystems of attractive surfaces for security intrusions, being vulnerable to multiple and simultaneous attacks. This research evaluates the performance of an Intrusion Detection System (IDS) configured with novel Machine Learning (ML) algorithms. This work presents a new balanced dataset (IDSAI) with intrusions generated under real-based attack settings. The experiments show that for the detection of intruders, the best algorithms are XGBoost, Gradient Boosting, Decision Tree, Random Forest, and Extra Trees. Which can generate predictions when trained and predicted with specific intrusions (i.e., ARP spoofing, Brute Force SSH, DDoS MAC Flood, ICMP echo request Flood, IP Fragmentation, SYN Flooding faster, SYN/ACK Flooding, SYN/ACK & RST Flooding, TCP Null, and UDP port scan), both of binary form (intrusion and no-intrusion) with up to 94 % of accuracy, as multiclass form (ten different intrusions and no-intrusion) with up to 92 % of accuracy. In contrast, up to 90 % of accuracy is achieved for prediction on the Bot-IoT dataset using models trained with the IDSAI dataset.

Citing

If you use our project for your research or if you find this paper and repository helpful, please consider citing the work.

Folders

• DBs This folder contain the datasets. Please unzip the Data.zip file when you download or clone the repository.
• Notebooks In this folder, you will find all algorithms and implementations for the different scenarios.

Requirements

This repository requires the following libraries:

• NumPy
• Pandas
• string
• Seaborn
• openpyxl
• scikit-learn
• keras-gpu
• scikit-image
• NLTK
• spaCy
• imbalanced-learn
• Matplotlib
• Yellowbrick
• Time
• OpenCV
• Pydot
• Graphviz
• dtreeviz
• Statsmodels
• XGBoost

This repository was developed in the Python 3 (3.8) programming language.

Package installation

If you don't use google colab, We highly recommend to use and install Python packages within an Anaconda environment. To create, execute the command below:

conda update -n base -c defaults conda

conda create --name NID python=3.8

pip install notebook

So, activate it:

conda activate NID

Packages installation

pip install ipykernel

and display of environment in jupyter

python -m ipykernel install --user --name NID --display-name "NID"

Now, install the libraries

conda install -c conda-forge matplotlib

conda install -c anaconda seaborn

conda install -c districtdatalabs yellowbrick

conda install -c conda-forge imbalanced-learn

pip install imblearn

conda install -c anaconda scikit-learn

pip install scikit-image

pip install anaconda keras-gpu

pip install opencv-python

pip install pandas

pip install pydot

conda install -c conda-forge spacy

conda install openpyxl

pip install graphviz

pip install seaborn

pip install dtreeviz

pip install statsmodels

pip install xgboost

pip install spacy

Execution

After installing all the Requirements, you must clone the repository using.

git clone https://github.com/BioAITeam/Intrusion-Detection-System-using-Machine-Learning.git

If you will use colab, upload the cloned folder to drive, then open the folder and run the notebook.

If you are going to use your computer, install:

conda install jupyter 

Enter the cloned folder, then enter the folder and run the notebook.