A Deep Reinforcement Learning Based Intrusion Detection System (DRL-IDS) for Securing Wireless Sensor Networks and Internet of Things

Benaddi, Hafsa; Ibrahimi, Khalil; Benslimane, Abderrahim; Qadir, Junaid

doi:10.1007/978-3-030-52988-8_7

Cited by 27 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Proposed schemes deployed in real-world settings will be subject to a highly variable behavior of the underlying network, making it unfeasible to be reproduced in a training dataset -demanding that the proposed scheme generalize the behavior of the network traffic [6], [7]. In practice, the network traffic behavior will change over time [34], [37], demanding model updates to be performed periodically. However, model updates are also a challenging task that often demands huge amounts of labeled training data to be provided.…”

Section: Discussionmentioning

confidence: 99%

“…The authors provided higher detection accuracy compared to traditional ML techniques while neglecting the challenges related to the actual deployment of their proposed scheme in production environments, such as the behavior changes in network traffic. Benaddi et al [37] proposed a RL scheme to detect malicious adversaries in wireless sensor networks. Their proposed approach was able to increase accuracy when compared to traditional ML techniques.…”

Section: Reinforcement Learning For Intrusion Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Reinforcement Learning for Intrusion Detection: More Model Longness and Fewer Updates

Santos

Viegas

Santin

et al. 2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

Several works have used machine learning techniques for network-based intrusion detection over the past few years. While proposed schemes have been able to provide high detection accuracies, they do not adequately handle the changes in network traffic behavior as time passes. Researchers often assume that model updates can be performed periodically as needed, although this is not easily feasible in real-world scenarios. This paper proposes a new intrusion detection model based on a reinforcement learning approach that aims to support extended periods without model updates. The proposal is divided into two strategies. First, it applies machine learning scheme as a reinforcement learning task to long-term learning -maintaining high reliability and high classification accuracies over time. Second, model updates are performed using a transfer learning technique coped with a sliding window mechanism that significantly decreases the need for computational resources and human intervention. Experiments performed using a new dataset spanning 8TB of data and four years of real network traffic indicate that current approaches in the literature cannot handle the evolving behavior of network traffic. Nevertheless, the proposed technique without periodic model updates achieves similar accuracy rates to traditional detection schemes implemented with semestral updates. In the case of performing periodic updates on our proposed model, it decreases the false positives up to 8%, false negatives up to 34%, with an accuracy variation up to only 6%, while demanding only seven days of training data and almost five times fewer computational resources when compared to traditional approaches.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Reinforcement Learning For Intrusion Detectionmentioning

confidence: 99%

Reinforcement Learning for Intrusion Detection: More Model Longness and Fewer Updates

Santos

Viegas

Santin

et al. 2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…O comportamento não estático do tráfego de rede foi também considerado por Ying-Feng Hsu e Morito Matsuoka [Benaddi et al 2020]. Os autores criaram uma mudanc ¸a de comportamento do tráfego de rede através da aplicac ¸ão de vários lotes do conjunto de dados de detecc ¸ão de intrusão enquanto avaliavam a sua técnica de acordo com cada lote.…”

Section: Trabalhos Relacionadosunclassified

Atualização de Modelo baseado em Aumento de Dados e Transferência de Aprendizagem para Detecção de Intrusão em Redes

Horchulhack

Viegas²,

Santin³

et al. 2022

Anais Do XXII Simpósio Brasileiro De Segurança Da Informação E De Sistemas Computacionais (SBSeg 2022)

View full text Add to dashboard Cite

Neste artigo apresentamos uma abordagem para atualização do modelo de aprendizagem de máquina para detecção de intrusão. Inicialmente, o tráfego de rede é aumentando por uma Redes Adversárias Generativas (GANs). Depois, as atualizações de modelos são realizadas por meio de Transferência de Aprendizagem sobre o conjunto de dados aumentado. O número de instâncias a ser rotuladas e os custos computacionais das atualizações do modelo são diminuídas significativa na proposta. A experimentação foi feita num conjunto de dados de 8TB (tráfego de rede de 1 ano), demonstrando a ineficiência dos trabalhos da literatura para detectar mudanças de comportamento no tráfego na rede. No caso do nosso modelo a taxa de falsos positivos diminuiu em até 18,1% quando atualizações periódicas são realizadas. As atualizações contemplaram somente 2,3% das instâncias, com uma diminuição de 14% no custo computacional.

show abstract

“…Referred publications Markov decision process [12,23,24,37,64,70,75,84,96,100,101,104,127,130,133,138,144,153,165,167,170,177,188,191,199], [203, 207, 211, 212, 214, 217, 220, 231, 252, 256-259, 263, 264, 272, 274, 281, 291, 309, 313, 320, 340, 343, 346], [369][370][371][372][373][374][375][376] Multiarmed bandit [61,66,102,198,351,377,378] Dynamic programming [16,19,27,52,68,70,84,90,93,…”

Section: Approachmentioning

confidence: 99%

The Applicability of Reinforcement Learning Methods in the Development of Industry 4.0 Applications

2021

View full text Add to dashboard Cite

Reinforcement learning (RL) methods can successfully solve complex optimization problems. Our article gives a systematic overview of major types of RL methods, their applications at the field of Industry 4.0 solutions, and it provides methodological guidelines to determine the right approach that can be fitted better to the different problems, and moreover, it can be a point of reference for R&D projects and further researches.

show abstract

A Deep Reinforcement Learning Based Intrusion Detection System (DRL-IDS) for Securing Wireless Sensor Networks and Internet of Things

Cited by 27 publications

References 18 publications

Reinforcement Learning for Intrusion Detection: More Model Longness and Fewer Updates

Reinforcement Learning for Intrusion Detection: More Model Longness and Fewer Updates

Atualização de Modelo baseado em Aumento de Dados e Transferência de Aprendizagem para Detecção de Intrusão em Redes

The Applicability of Reinforcement Learning Methods in the Development of Industry 4.0 Applications

Contact Info

Product

Resources

About