Intrusion Detection Model for IoT Using Recurrent Kernel Convolutional Neural Network

Kumar, C. U. Om; Marappan, Suguna; Murugeshan, Bhavadharini; Beaulah, V. Mercy Rajaselvi

doi:10.1007/s11277-022-10155-9

Cited by 17 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compared the DL-BiLSTM model with recent studies ( Shieh et al, 2021 ; Wang et al, 2021 ; Bhardwaj, Mangat & Vig, 2020 ; Qazi, Almorjan & Zia, 2022 ; Qureshi et al, 2021 ; Alharbi et al, 2021 ; Attique, Hao & Ping, 2022 ; Om Kumar et al, 2022 ) on two datasets, CIC IDS2017 and N-BaIoT, in order to further illustrate the superiority of the detection performance of the model provided in this article. The experimental results can be seen in Tables 7 and 8 , and the performance test results not given in the paper are indicated with “-” in the table.…”

Section: Experimental Results and Analysismentioning

confidence: 99%

A lightweight intrusion detection method for IoT based on deep learning and dynamic quantization

Wang,

Chen,

Yang

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

Intrusion detection ensures that IoT can protect itself against malicious intrusions in extensive and intricate network traffic data. In recent years, deep learning has been extensively and effectively employed in IoT intrusion detection. However, the limited computing power and storage space of IoT devices restrict the feasibility of deploying resource-intensive intrusion detection systems on them. This article introduces the DL-BiLSTM lightweight IoT intrusion detection model. By combining deep neural networks (DNNs) and bidirectional long short-term memory networks (BiLSTMs), the model enables nonlinear and bidirectional long-distance feature extraction of complex network information. This capability allows the system to capture complex patterns and behaviors related to cyber-attacks, thus enhancing detection performance. To address the resource constraints of IoT devices, the model utilizes the incremental principal component analysis (IPCA) algorithm for feature dimensionality reduction. Additionally, dynamic quantization is employed to trim the specified cell structure of the model, thereby reducing the computational burden on IoT devices while preserving accurate detection capability. The experimental results on the benchmark datasets CIC IDS2017, N-BaIoT, and CICIoT2023 demonstrate that DL-BiLSTM surpasses traditional deep learning models and cutting-edge detection techniques in terms of detection performance, while maintaining a lower model complexity.

show abstract

Section: Experimental Results and Analysismentioning

confidence: 99%

A lightweight intrusion detection method for IoT based on deep learning and dynamic quantization

Wang,

Chen,

Yang

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Kumar et al [74] have proposed the IDS using recurrent kernel CNN and Modified Monarch Butterfly Optimization. They used the min-max technique for pre-processing and improved battle royale optimization for the extraction of optimal features.…”

Section: Related Workmentioning

confidence: 99%

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

Sana,

Nazir,

Yang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Background and Motivation:The ever-expanding Internet of Things (IoT) landscape presents a double-edged sword. While it fosters interconnectedness, the vast amount of data generated by IoT devices creates a larger attack surface for cybercriminals. Intrusions in these environments can have severe consequences. To combat this growing threat, robust intrusion detection systems (IDS) are crucial. The data comprised by this attack is multivariate, highly complex, non-stationary, and nonlinear. To extract the complex patterns from this complex data, we require the most robust, optimized tools. Methods: Machine learning (ML) and deep learning (DL) have emerged as powerful tools for IDSs, offering high accuracy in detecting and preventing security breaches. This research delves into anomaly detection, a technique that identifies deviations from normal system behavior, potentially indicating attacks. Given the complexity of anomaly data, we explore methods to improve detection performance. Proposed Approach: This research investigates the design and evaluation of a novel IDS. We leverage and optimize supervised ML methods like tree-based Support Vector Machines (SVM), ensemble methods, and neural networks (NN) alongside the cutting-edge DL approach of long short-term memory (LSTM) and vision transformers (ViT). We optimized the hyperparameters of these algorithms using a robust Bayesian optimization approach. Results: The implemented ML models achieved impressive training accuracy, with Random Forest and Ensemble Bagged Tree surpassing 99.90% of accuracy, an AUC of 1.00, an F1-score, and a balanced Matthews Correlation Coefficient (MCC) of 99.78%. While the initial deep learning LSTM model yielded an accuracy of 99.97%, the proposed ViT architecture significantly boosted performance with 100% of all metrics, along with a validation accuracy of 78.70% and perfect training accuracy. Conclusion: Our findings demonstrate the potential of the proposed methods for enhanced attack and intrusion detection. This improved detection capability can be instrumental in safeguarding system integrity, identifying fraudulent activity, and optimizing system performance in IoT networks.

show abstract

“…66 % of the global population will have smartphones by 2020 (Data Reportal, 2021) where 93 million used Android based mobile devices and a perennial increase of 1,8 %. (10) In fact, it has been used in a broad range of devices, including wearable's such as augmented reality headsets and smart watches, tablets, smartphones, and wearable's. Because the Android operating system is widely used, it is open-source, and we frequently save sensitive data on our mobile devices, rogue code writers write increasingly aggressive scripts every day with the intention of stealing our data.…”

Section: Introductionmentioning

confidence: 99%

Ensemble classification based hybrid dual-channel convolution neural network (dccnn) with enhanced manta ray foraging optimization (emrfo) algorithm for cyber security malware threats detection

Vijayalakshmi,

Karthika

2024

Salud, Ciencia y Tecnología - Serie de Conferencias

View full text Add to dashboard Cite

Introduction: This study suggests usage of hybrid deep learning (DL) for identifying malwares in Internet of Things (IoT) networks. Furthermore, Channel Boost STM-RENet (CB-STM-RENet) is proposed as a DCCNN optimization technique that extends the split-change-merge model. Malware detection is performed using Hybrid Dual Channel Convolutional Neural Network (DCCNN) and Manta Ray Forage Optimization. Methods: In this context, introduce a single-block convolutional STM known as DCCNN in CB-STM-RENet that performs local and spatial processing at the same time. The systematic use of the region and the deployment of parallel socialization processes facilitate the investigation of the unity of the region, the diversity of forces and the defining characteristics of the region. Three versions of DL: STM-RENet, DenseNet201 and InceptionResNetV2 (IRNV2) are proposed which work together to optimize DCCNN using split-change-merge in a unique way to improve generalization Hybrid learning. This dataset is a Google Code Jam (GCJ) for IoT malware detection challenges. Results: The experimental results of the suggested method are better than existing methods for obtained accuracies and values of precision, specificity, F1 scores, MCC, and avg. processing times in classifications of cyber threats.

show abstract

Intrusion Detection Model for IoT Using Recurrent Kernel Convolutional Neural Network

Cited by 17 publications

References 22 publications

A lightweight intrusion detection method for IoT based on deep learning and dynamic quantization

A lightweight intrusion detection method for IoT based on deep learning and dynamic quantization

Securing the IoT Cyber Environment: Enhancing Intrusion Anomaly Detection With Vision Transformers

Ensemble classification based hybrid dual-channel convolution neural network (dccnn) with enhanced manta ray foraging optimization (emrfo) algorithm for cyber security malware threats detection

Contact Info

Product

Resources

About