Intrusion Detection System in Ad Hoc Networks with Artificial Neural Networks and Algorithm K-Means

Canêdo, Daniel Rosa; Romariz, Alexandre Ricardo Soares

doi:10.1109/tla.2019.8931198

Cited by 16 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most existing algorithms simply categorise data using the trained model, which might lead to suboptimal detection results because of the inherent biases in the model. In this work, k-NN classifier [37] is used which is based on the ratio of data distribution in order to address the aforementioned issue. The k-NN approach for classifying data essentially determines the category of test data by determining which of the training set's k closest neighbours occurs more often.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

“…Blackhole Attack This work focuses is on securing networks against network attacks and chooses blackhole attack optimization as an example case for our experiments in an IoT environment. In [39] is an optimisation algorithm used in conjunction with the ANN [37] method to discover these malicious nodes. This optimization alogirthm includes following steps:…”

Section: Enhancement Using Detection Modulementioning

confidence: 99%

“…For the purpose of identifying and categorising attacks in computer networks, the authors of [37] propose a two-stage Ad Hoc Wireless Intrusion Detection System. The accuracy has been calculated in [37] as 97%.…”

Section: Enhancement Using Detection Modulementioning

confidence: 99%

See 2 more Smart Citations

Neural Networks and Blackhole Optimization in a Cooperative IDS for IoT

Li,

Liu,

Jia

2024

Preprint

View full text Add to dashboard Cite

In this paper, network security issues related to the Internet of Things (IoT) are discussed, with an emphasis on grey- and black-hole attacks (BHAs and GHAs). IoT systems lack security safeguards like intrusion detection systems because of their decentralised nature. The study suggests a defence mechanism against BHA utilising an attack selection algorithm and an optimisation strategy using a neural network to handle this. The book also emphasises how challenging it is to implement cutting-edge security features on low-power IoT devices. The suggested method comprises creating a training set and teaching a neural network for classification. This method is specifically designed for IoT intrusion detection systems. In order to increase overall efficiency, the article’s findings section explores how to improve system performance by choosing the right nodes for data transmission.

show abstract

Section: Artificial Neural Networkmentioning

confidence: 99%

Section: Enhancement Using Detection Modulementioning

confidence: 99%

See 1 more Smart Citation

Neural Networks and Blackhole Optimization in a Cooperative IDS for IoT

Li,

Liu,

Jia

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…With the increasing complexity of the network environment, detection systems have gradually shifted from rule-and expert-based methods to machine-learning-based methods, such as decision trees [3], support vector machines [4], XGBoost [5], neural networks [6], and so on. XGBoost, as one of the supervised learning models, still has unique advantages in the scenario of limited training samples, short training time, and lack of parameter tuning knowledge when neural networks, especially deep neural networks, are becoming more and more popular.…”

Section: Introductionmentioning

confidence: 99%

Intrusion Detection Systems Based on Logarithmic Autoencoder and XGBoost

Fan

2022

Security and Communication Networks

View full text Add to dashboard Cite

An intrusion detection system (IDS) is a network security device that performs real-time monitoring of network transmissions and sends out alarms or takes active response measures when suspicious transmissions are found. In this regard, many researches have combined traditional machine learning models with other optimization algorithms to improve intrusion detection performance. However, although the existing intrusion detection model can effectively improve the performance of the model, there are still problems such as unsatisfactory detection accuracy and data preprocessing operations that may lead to a decrease in accuracy. To solve this problem, in this paper, we have proposed a novel intrusion detection system model based on logarithmic autoencoder (LogAE) and eXtreme Gradient Boosting (XGBoost). First, we build LogAE to learn the hidden features of the input data to reconstruct new data similar to the training samples, with the purpose of highlighting important features. It is worth mentioning that LogAE is not necessary to normalize the training data. This is because we add a logarithmic layer to learn this mapping. Then, XGBoost is used as a classifier to identify the data set that combines the original data set with the generated data set. In the experiment, our proposed model is evaluated on the UNSW-NB15 data set and CICIDS2017 data set. Additionally, we use accuracy, recall, precision, F1-score, and runtime as evaluation metrics. For detection performance, the detection accuracy of our proposed model is 95.11% for UNSW-NB15 and 99.92% for CICIDS2017, which is better than most state-of-the-art intrusion detection methods. Meantime, the runtime of our proposed model is the lowest for UNSW-NB15.

show abstract