LocKedge: Low-Complexity Cyberattack Detection in IoT Edge Computing

Hương, Trương Thu; Bac, Ta Phuong; Long, Dao Minh; Thang, Bui Doan; Binh, Nguyen Thi My; Luong, Tran Duc; Phuc, Tran Kim

doi:10.1109/access.2021.3058528

Cited by 53 publications

(40 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [ 40 ], the Convolutional Neural Network (CNN) model outperformed the RNN, LSTM and GRU models. Huong et al [ 41 , 42 ] proposed a low-complexity edge-cloud DNN model, which achieved better performance than the kNN, DT, RF and SVM models. Lee et al [ 43 ] employed RF for botnet attack classification in an IoT smart factory.…”

Section: Review Of Related Workmentioning

confidence: 99%

SMOTE-DRNN: A Deep Learning Algorithm for Botnet Detection in the Internet-of-Things Networks

Popoola

Adebisi

Ande

et al. 2021

Sensors

View full text Add to dashboard Cite

Nowadays, hackers take illegal advantage of distributed resources in a network of computing devices (i.e., botnet) to launch cyberattacks against the Internet of Things (IoT). Recently, diverse Machine Learning (ML) and Deep Learning (DL) methods were proposed to detect botnet attacks in IoT networks. However, highly imbalanced network traffic data in the training set often degrade the classification performance of state-of-the-art ML and DL models, especially in classes with relatively few samples. In this paper, we propose an efficient DL-based botnet attack detection algorithm that can handle highly imbalanced network traffic data. Specifically, Synthetic Minority Oversampling Technique (SMOTE) generates additional minority samples to achieve class balance, while Deep Recurrent Neural Network (DRNN) learns hierarchical feature representations from the balanced network traffic data to perform discriminative classification. We develop DRNN and SMOTE-DRNN models with the Bot-IoT dataset, and the simulation results show that high-class imbalance in the training data adversely affects the precision, recall, F1 score, area under the receiver operating characteristic curve (AUC), geometric mean (GM) and Matthews correlation coefficient (MCC) of the DRNN model. On the other hand, the SMOTE-DRNN model achieved better classification performance with 99.50% precision, 99.75% recall, 99.62% F1 score, 99.87% AUC, 99.74% GM and 99.62% MCC. Additionally, the SMOTE-DRNN model outperformed state-of-the-art ML and DL models.

show abstract

Section: Review Of Related Workmentioning

confidence: 99%

SMOTE-DRNN: A Deep Learning Algorithm for Botnet Detection in the Internet-of-Things Networks

Popoola

Adebisi

Ande

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Edge computing has become an essential approach in IoT-based networks to store data on the cloud to reduce cloud server workload and real-time event detection [32]. Lockedge [33] proposed an edge cloud infrastructure that fulfills the edge layer's detection requirements to ensure a quick response. An energy-efficient Q-Learning-based data aggregation protocol is proposed for IoT to save energy.…”

Section: Iintroductionmentioning

confidence: 99%

An Energy-Efficient Data Aggregation Mechanism for IoT Secured by Blockchain

et al. 2022

View full text Add to dashboard Cite

The Internet of Things (IoT) is getting important and interconnected technologies of the world, consisting of sensor devices. The internet is smoothly changing from an internet of people towards an Internet of Things, which permits various objects to connect to another wirelessly. The energy consumption of the IoT routing protocol can affect the network life span. In addition, the high volume of data produced by IoT will result in transmission collision, security issues, and energy dissipation due to increased data redundancy because tiny sensors are usually hard to recharge after they are deployed. Generally, to save energy, data aggregation reduces data redundancy at each node by turning some nodes into sleep mode and others into wake mode. Therefore, it is important to group the nodes with high data similarity using the fuzzy matrix. Then, the data received from the member nodes at the Cluster Head (CH) are analyzed using a fuzzy similarity matrix for clustering. In the next step, after clustering, some nodes are chosen from all groups as redundant nodes. The sleep scheduling mechanism is then applied to reduce data redundancy, network traffic jamming, and transmission costs. We have proposed an Energy-Efficient Data Aggregation Mechanism (EEDAM) secured by blockchain, which uses a data aggregation mechanism at the cluster level to save energy. As edge computing is used to provide on-demand trusted services to IoT with minimum delay, blockchain is integrated inside a cloud server, so the edge is validated by the blockchain to provide secure services to IoT. Finally, we performed simulations to calculate the performance of the proposed mechanism and compared it with the conventional energy-efficient algorithms. The simulation results show that the proposed structural design can successfully reduce the amount of data, provide proper security to the IoT, and extend the wireless sensor network (WSN).

show abstract

“…Each IoT application may have different Quality of Service (QoS) requirements. To realize a reliable and robust IoT network, meeting the QoS requirements is needed [16][17][18][19][20][21][22][23][24][25].…”

Section: Introductionmentioning

confidence: 99%

Reliable Internet of Things: Challenges and Future Trends

et al. 2021

View full text Add to dashboard Cite

The Internet of Things (IoT) is a vital component of many future industries. By intelligent integration of sensors, wireless communications, computing techniques, and data analytics, IoT can increase productivity and efficiency of industries. Reliability of data transmission is key to realize several applications offered by IoT. In this paper, we present an overview of future IoT applications, and their major communication requirements. We provide a brief survey of recent work in four major areas of reliable IoT including resource allocation, latency management, security, and reliability metrics. Finally, we highlight some of the important challenges for reliable IoT related to machine learning techniques, 6G communications and blockchain based security that need further investigation and discuss related future directions.

show abstract

LocKedge: Low-Complexity Cyberattack Detection in IoT Edge Computing

Abstract: and Technology (HUST) under project number T2020-SAHEP-010.

Cited by 53 publications

References 21 publications

SMOTE-DRNN: A Deep Learning Algorithm for Botnet Detection in the Internet-of-Things Networks

SMOTE-DRNN: A Deep Learning Algorithm for Botnet Detection in the Internet-of-Things Networks

An Energy-Efficient Data Aggregation Mechanism for IoT Secured by Blockchain

Reliable Internet of Things: Challenges and Future Trends

Contact Info

Product

Resources

About