Attack and Anomaly Detection in IoT Networks using Machine Learning

Abstract

For quite a few years now the name Internet of Things (IoT) has been around. IoT is a technology capable of revolutionizing our way of life, in sectors ranging from transportation to health, from entertainment to our interactions with government. Even this great opportunity presents a number of critical obstacles. As we strive to develop policies, regulations, and governance that form this development without stifling creativity, the increase in the number of devices and the frequency of that increase presents problems to our security and freedom. This work attentions on the security aspect of IoT networks by examining the serviceability of machine learning algorithms in detecting anomaliesthat are contained within such network data. It discusses (Machine Learning (ML) algorithms which are used effectively in relatively similar situations and compares them using several parameters and methods. The following algorithms are implemented in this work: Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM), and Decision tree Algorithm. The Random Forest algorithm obtained the best results, with an accuracy of 99.5 per cent.

A Machine Learning-Based Intrusion Detection System for IoT Electric Vehicle Charging Stations (EVCSs)

“…The random forest algorithm achieved the best accuracy of 99.5% among the others. While the methodology and the data samples were different from our study, as we used highly randomized data, the results for the above study, along with the other study of Thamataiselvi et al [15], was in line with our results too.…”

Application of Machine Learning Algorithms for the Validation of a New CoAP-IoT Anomaly Detection Dataset

“…IoT-NID [25] Smart home, laptops, and smartphones Mirai, MITM, DoS, scanning, etc Autoencoder [28] IoT-23 [26] Smart home (real IoT devices) Mirai, Torii, and Gafgyt RF, NB, SVM, DT [27], autoencoder [28] ToN-IoT [30] Weather and industrial control sensors (MQTT)…”

HH-NIDS: Heterogeneous Hardware-Based Network Intrusion Detection Framework for IoT Security

“…They have achieved up to 99% accuracy by applying the random forest (RF) classifier; however, the dataset used in this research was small (357,952 records) compared to other practical datasets. RF has also been claimed, in [22], as the best algorithm for detecting anomalies. Meanwhile, DL has shown better performance when dealing with large datasets.…”

“…They have improved this result to 99.99% by applying LSTM in [49]; however, F1-score is not mentioned in this research. Authors in [22,34] have reported RF as the best algorithm for detecting attacks with 100% and 99.50% accuracy, respectively. Authors in [37] have achieved accuracy with only 0.14% greater than our NN model.…”