Securing IoT Devices: A Robust and Efficient Deep Learning with a Mixed Batch Adversarial Generation Process for CAPTCHA Security Verification

Dankwa, Stephen; Yang, Lu

doi:10.3390/electronics10151798

Cited by 5 publications

(5 citation statements)

References 73 publications

(61 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some studies [ 6 , 15 ] suggest the use of the ensemble learning approach in IDS contexts generally. However, limited consideration has been given to such an approach in the IoT context [ 37 ]. Moreover, few works employ ensemble defense methods that primarily focus on modifying the model itself without consideration of the other defense approaches [ 4 ].…”

Section: Background and Related Workmentioning

confidence: 99%

RobEns: Robust Ensemble Adversarial Machine Learning Framework for Securing IoT Traffic

Alkadi,

Al-Ahmadi,

Ben Ismail

2024

Sensors

View full text Add to dashboard Cite

Recently, Machine Learning (ML)-based solutions have been widely adopted to tackle the wide range of security challenges that have affected the progress of the Internet of Things (IoT) in various domains. Despite the reported promising results, the ML-based Intrusion Detection System (IDS) proved to be vulnerable to adversarial examples, which pose an increasing threat. In fact, attackers employ Adversarial Machine Learning (AML) to cause severe performance degradation and thereby evade detection systems. This promoted the need for reliable defense strategies to handle performance and ensure secure networks. This work introduces RobEns, a robust ensemble framework that aims at: (i) exploiting state-of-the-art ML-based models alongside ensemble models for IDSs in the IoT network; (ii) investigating the impact of evasion AML attacks against the provided models within a black-box scenario; and (iii) evaluating the robustness of the considered models after deploying relevant defense methods. In particular, four typical AML attacks are considered to investigate six ML-based IDSs using three benchmarking datasets. Moreover, multi-class classification scenarios are designed to assess the performance of each attack type. The experiments indicated a drastic drop in detection accuracy for some attempts. To harden the IDS even further, two defense mechanisms were derived from both data-based and model-based methods. Specifically, these methods relied on feature squeezing as well as adversarial training defense strategies. They yielded promising results, enhanced robustness, and maintained standard accuracy in the presence or absence of adversaries. The obtained results proved the efficiency of the proposed framework in robustifying IDS performance within the IoT context. In particular, the accuracy reached 100% for black-box attack scenarios while preserving the accuracy in the absence of attacks as well.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

RobEns: Robust Ensemble Adversarial Machine Learning Framework for Securing IoT Traffic

Alkadi,

Al-Ahmadi,

Ben Ismail

2024

Sensors

View full text Add to dashboard Cite

show abstract

“…The researchers in [66] outlined an MBAGP-CNN system that incorporates vulnerability verification and defense techniques using deep learning with a mixed batch adversarial generation process. Crafting adversarial text-based CAPTCHAs is performed using several attack methods such as FGSM [26], Iterative Fast Gradient Sign Method (I-FGSM) [36], and the Momentum Iterative Fast Gradient Sign Method (MI-FGSM) [66] algorithms. Specifically, the system works on breaking the transferability attack by feeding both the original and the adversarial CAPTCHA images to the CNN model to perform the retraining process.…”

Section: Applications Of Aml Towards Internet Of Things (Iot) Robustnessmentioning

confidence: 99%

Better Safe Than Never: A Survey on Adversarial Machine Learning Applications towards IoT Environment

Alkadi,

Al-Ahmadi,

Ismail

2023

Applied Sciences

View full text Add to dashboard Cite

Internet of Things (IoT) technologies serve as a backbone of cutting-edge intelligent systems. Machine Learning (ML) paradigms have been adopted within IoT environments to exploit their capabilities to mine complex patterns. Despite the reported promising results, ML-based solutions exhibit several security vulnerabilities and threats. Specifically, Adversarial Machine Learning (AML) attacks can drastically impact the performance of ML models. It also represents a promising research field that typically promotes novel techniques to generate and/or defend against Adversarial Examples (AE) attacks. In this work, a comprehensive survey on AML attack and defense techniques is conducted for the years 2018–2022. The article investigates the employment of AML techniques to enhance intrusion detection performance within the IoT context. Additionally, it depicts relevant challenges that researchers aim to overcome to implement proper IoT-based security solutions. Thus, this survey aims to contribute to the literature by investigating the application of AML concepts within the IoT context. An extensive review of the current research trends of AML within IoT networks is presented. A conclusion is reached where several findings are reported including a shortage of defense mechanisms investigations, a lack of tailored IoT-based solutions, and the applicability of the existing mechanisms in both attack and defense scenarios.

show abstract

“…Similarly, categories for DL techniques include unsupervised, supervised, and hybrid approaches. (11,12) . The fundamental benefit of deep learning over traditional machine learning is that it performs much better on large datasets.…”

Section: And ML For Iot Security and Privacymentioning

confidence: 99%

“…In IoT environments, DL approaches have been devised for signal authentication (11,12) . The LSTM architecture is used to extract an array of features from inputs of IoT device.…”

Section: Application Of DL In Iot Securitymentioning

confidence: 99%

“…These assumptions are not realistic in most of the times. LSTM algorithm with game theory techniques for a mixed Nash balance (NE) approach has been adapted to address the security problem in IoT network (11,12) . Although the implantation complexity of game theory and the higher computation time are claimed to be taken care by the constituent counterparts of NE approach, the mitigation of either difficulties are not fully overcome.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Implementation of Machine Learning and Deep Learning for Securing the Devices in IOT Systems

Jagannath¹,

Mohite²,

Gupta³

et al. 2023

IJST

View full text Add to dashboard Cite

Objectives: To fix the security issues of devices in internet of things (IOT) systems that arise when Machine learning (ML) and Deep Learning algorithm (DLA) are implemented in the IOT systems. Methods: Each packet of IoT threats has been filtered by using suitable attack model for primary attack detection. In deep learning, network traffic field's information is extracted from of packet and is used as the training features. Attack is proposed to be detected by the use of database-based features of attack. If attack is found then packet will be discarded along with update feedback to filtering stage for primary attack detection. Knowledge Discovery and Data Mining (KDD) dataset is used in this study. Different types of Mirai attack are considered for classification. The performance of the proposed method has been compared in terms of accuracy and execution time with earlier work using traditional ML-DL methods. Findings: The proposed technique is seen to achieve high accuracy level at the least computational time, concurrently with much higher recall and G-mean values Several algorithms are used to secure the IOT devices from various types of threats and a comparison is depicted with their accuracy and execution time. Results show that the proposed methodology using Convolution neural network (CNN) for classification, can achieve the accuracy ~0.9976 with execution time1.30 sec and G-mean 0.7865 only. Novelty: The proposed method is essentially a judiciously configured ML-DL technique which is novel and exhibits better performance in terms of mitigating IoT threats.

show abstract

Securing IoT Devices: A Robust and Efficient Deep Learning with a Mixed Batch Adversarial Generation Process for CAPTCHA Security Verification

Cited by 5 publications

References 73 publications

RobEns: Robust Ensemble Adversarial Machine Learning Framework for Securing IoT Traffic

RobEns: Robust Ensemble Adversarial Machine Learning Framework for Securing IoT Traffic

Better Safe Than Never: A Survey on Adversarial Machine Learning Applications towards IoT Environment

Implementation of Machine Learning and Deep Learning for Securing the Devices in IOT Systems

Contact Info

Product

Resources

About