Search

Using deep learning models to deal with the classification tasks in network traffic offers a new approach to address the imbalanced Internet of Things malicious traffic classification problems. However, the employment difficulty of these models may be immense due to their high resource consumption and inadequate interpretability. Fortunately, the effectiveness of sampling methods based on the statistical principles in imbalance data distribution indicates the path. In this paper, we address these challenges by proposing a hybrid sampling method, termed HSS, which integrates undersampling and oversampling techniques. Our approach not only mitigates the imbalance in malicious traffic but also fine-tunes the sampling threshold to optimize performance, as substantiated through validation tests. Employed across three distinct classification tasks, this method furnishes simplified yet representative samples, enhancing the baseline models' classification capabilities by a minimum of 6.02% and a maximum of 182.66%. Moreover, it notably reduces resource consumption, with sample numbers diminishing to a ratio of at least 83.53%. This investigation serves as a foundation, demonstrating the efficacy of HSS in bolstering security measures in IoT networks, potentially guiding the development of more adept and resource-efficient solutions. [ABSTRACT FROM AUTHOR]

This paper introduces a sophisticated approach to network security, with a primary emphasis on utilizing deep learning for intrusion detection. In real-world scenarios, the high dimensionality of training data poses challenges for simple deep learning models and can lead to vanishing gradient issues with complex neural networks. Additionally, uploading network traffic data to a central server for training raises privacy concerns. To tackle these issues, the paper introduces a Residual Network (ResNet)-based deep learning model trained using a federated learning approach. The ResNet effectively tackles the vanishing gradient problem, while federated learning enables multiple Internet Service Providers (ISPs) or clients to engage in joint training without sharing their data with third parties. This approach enhances accuracy through collaborative learning while maintaining privacy. Experimental results on the X-IIoTID dataset indicate that the proposed model outperforms conventional deep learning and machine learning methods in terms of accuracy and other metrics used for evaluation. Specifically, the proposed methodology achieved 99.43% accuracy in a centralized environment and 99.16% accuracy in a federated environment. [ABSTRACT FROM AUTHOR]

The proliferation of Internet of Things (IoT) devices has revolutionized various domains, but it has also brought forth numerous security challenges. One of the most concerning threats is the emergence of hybrid attacks, which combine multiple attack vectors to exploit vulnerabilities in IoT networks. Existing security mechanisms often struggle to effectively predict and detect these sophisticated hybrid attacks, leading to compromised system integrity and data confidentiality. In this paper, we propose robust learning approach, named HybridRobustNet (HRN), for predicting and detecting hybrid attacks over IoT networks. HRN integrates machine learning algorithms, deep neural networks, and ensemble techniques to achieve enhanced detection accuracy and resilience against evolving hybrid attack patterns. By leveraging a diverse set of features, including network traffic patterns, device behavior, and communication characteristics, HRN effectively captures the complex relationships and dependencies between various attack components. Furthermore, the proposed approach incorporates real-time adaptive learning mechanisms, enabling it to dynamically adapt to new attack strategies and mitigate false positives. To evaluate the effectiveness of HRN, extensive experiments were conducted on a realistic IoT testbed comprising heterogeneous devices and attack scenarios. The results demonstrate that HRN outperforms state-of-the-art approaches in terms of attack detection accuracy, robustness against evasion techniques, and low false positive rates. Additionally, its computational efficiency makes it suitable for deployment in resource-constrained IoT environments. The contributions of this work are twofold. Firstly, it addresses the pressing need for robust detection mechanisms against hybrid attacks, which can have severe consequences for IoT networks. Secondly, it introduces a unique and adaptive learning approach, HRN, which exhibits superior performance and adaptability in the face of emerging attack strategies. The findings presented in this article provide valuable insights into the design of effective security mechanisms for IoT networks and pave the way for future research in the field of hybrid attack detection. [ABSTRACT FROM AUTHOR]

With rapid development and deployment of artificial intelligence methods, the Industrial Internet of Things (IIoT) has highly developed to fast tracing industrial communications and optimizing manufacturing processes. In Industry 5.0, hyper-automation process as one of technological trends navigates industrial entities to intelligent devices of the IIoT, cloud computing, smart robotics, smart agile software and embedded components by high complexity and reliability. By increasing data communication in the IIoT environments and cloud computing, the security and safety of hyper-automation process is also increasingly unstable and challengeable with respect to cyber-attacks, unstructured malwares and abnormal activities. With the diversification and unexpected behaviors of cyber-security threats, traditional cyber-attack detection systems have some critical problems with increasing massive data including unappropriated feature selection and extraction, high computation time in prediction and inaccurate classification models. Due to the above-mentioned challenges, this paper presents a new cloud-based cyber-attack detection architecture based on Ensemble Bagged Trees Detection (EBTD) algorithm for predicting malicious behaviors and types of cyber-attacks for hyper-automation process in the IIoT. The proposed architecture uses Analysis of Variance (ANOVA) and a priority-based feature selection and extraction model to find the optimal features with highly dependent on the network traffic, computation time, malicious behaviors, and types of attacks. Then, experimental results are conducted using the technical data sets UNSW-NB15 and NSL-KDD. The simulation results show that the proposed architecture performance performs better than other case studies and prediction models and effective on optimization of large-scale cyber-attack detection systems for critical hyper-automation process in the IIoT environment. [ABSTRACT FROM AUTHOR]

Sensors, wearables, mobile devices, and other Internet of Things (IoT) devices are becoming increasingly integrated into all aspects of our lives. They are capable of gathering enormous amounts of data, such as image data, which can then be sent to the cloud for processing. However, this results in an increase in network traffic and latency. To overcome these difficulties, edge computing has been proposed as a paradigm for computing that brings processing closer to the location where data is produced. This paper explores the merging of cloud and edge computing for IoT and investigates approaches using machine learning for dimensionality reduction of images on the edge, employing the autoencoder deep learning-based approach and principal component analysis (PCA). The encoded data is then sent to the cloud server, where it is used directly for any machine learning task without significantly impacting the accuracy of the data processed in the cloud. The proposed approach has been evaluated on an object detection task using a set of 4000 images randomly chosen from three datasets: COCO, human detection, and HDA datasets. Results show that a 77% reduction in data did not have a significant impact on the object detection task's accuracy. [ABSTRACT FROM AUTHOR]

In the cyberspace, device identification has become one of the most important factors in improving security of a network, containing both Internet of Things (IoT) and non-IoT devices. Resource-constraint IoT devices are generally more vulnerable than non-IoT devices, to different kinds of security threats, including Mirai botnet and spoofing attacks. In this paper, a device fingerprinting (DFP) scheme has been proposed based on the analysis of network traffic characteristics. Four statistical features from two device-specific features have been selected using statistical assessment to generate DFP for classification task using a supervised machine learning Random Forest classifier. Experimental results have shown that the proposed DFP scheme is able to classify device type with 99.81% accuracy on the public UNSW dataset, whilst accuracies of 99.50% and 97.10% have been reported for the identification of individual IoT and non-IoT devices, respectively. The proposed DFP scheme has also demonstrated superior performance as compared to other DFP methods in the literature, despite using less number of features and packets for DFP. These signify that the proposed DFP scheme can be used as a network security reinforcement tool in a heterogeneous network environment. [ABSTRACT FROM AUTHOR]

The rapid growth of Internet of Things (IoT) devices has changed human interactions with the environment. IoT networks require specialized defense strategies distinct from traditional corporate contexts. Security measures such as anti-malware software, firewalls, authentication protocols, and encryption techniques are established but face limitations against evolving attack strategies. Therefore, this study proposes an intrusion detection approach for a realistic IoT environment, employing various variants of deep learning models such as deep neural networks (DNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs). The research tested three variants for each model: DNN1, DNN2, DNN3, CNN1, CNN2, CNN3, RNN1, RNN2, RNN3. All these variants are customized and tuned differently to analyze the efficacy of the suggested methodology. Likewise, notable observation highlights the significance of aligning training and validation accuracy curves that indicate controlled overfitting, validating the model's reliability in accurately predicting intrusion and benign network traffic in IoT settings. Results reveal that RNN1 achieves the best results: accuracy of 98.61%, precision of 98.55%, recall of 98.61%, and F1-score of 98.57% compared with other DNN and CNN architectures and benchmark papers. This study advances intrusion detection within IoT networks through a comprehensive evaluation of deep learning models and inspires ongoing research to enhance intrusion detection systems' resilience in dynamic IoT environments. [ABSTRACT FROM AUTHOR]

Securing confidential data against unauthorized users leads to access control policies with the rapid progression of Internet of Things (IoT) devices. Because of high mobility subscribers, the dynamic IoT environment is subjected to high signaling overhead which remains a challenging issue to guarantee data dissemination to legitimate users. The group's key management schemes are the central mechanism to deal with dynamic environments. But they are centralized concepts that cause scalability issues and suffer in handling large numbers of subscribers. Therefore, this paper proposes a modified deep reinforcement oppositional wolf learning-based group key management (MDROWL-GKM) system to monitor the data obtained in IoT properly. It does not maximize the network traffic as well as computational overhead when a group member leaves or joins. With the inclusion of an opposition-based learning gray wolf optimization algorithm, the overload issue of the modified deep reinforcement method is eliminated and the performance is enhanced. The efficacy of the proposed MDROWL-GKM system is investigated using different measures namely storage overhead, computation overhead, access response time, space complexity, re-evaluation time, policy adjustment accuracy, and communication overhead. The experimental analysis proves that the proposed MDROWL-GKM system is superior to other state-of-the-art techniques, particularly with high policy adjustment accuracy (96%), less communication overhead (8 μs) and area under curve (AUC) rate (0.982). [ABSTRACT FROM AUTHOR]

The increasing prevalence of Internet of Things (IoT) systems has made them attractive targets for malicious actors. To address the evolving threats and the growing complexity of detection, there is a critical need to search for and develop new algorithms that are fast and robust in detecting and classifying dangerous network traffic. In this context, deep reinforcement learning (DRL) is gaining recognition as a prospective solution in numerous fields as it enables autonomous agents to cooperate with their environment for decision-making without relying on human experts. This article presents an innovative approach to intrusion detection in IoT systems using an adversarial reinforcement learning (RL) algorithm known for its exceptional predictive capabilities. The predictive process relies on a classifier, implemented as a streamlined and highly efficient neural network. Embedded within this classifier is a policy function meticulously trained using an innovative RL model. Importantly, this model ensures that the environment's behavior is dynamically fine-tuned simultaneously with the learning process, improving the overall effectiveness of the intrusion detection approach. The efficiency of our proposal was assessed using the Bot-IoT database, consisting of a mixture of legitimate IoT network traffic and simulated attack scenarios. Our scheme shows superior performance compared to existing ones. Therefore, our approach to IoT intrusion detection can be considered a valuable alternative to existing methods, capable of significantly improving the IoT systems' security. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) has shown incredible adaptability in recent years and has become an integral part of human life. The proliferation of IoT technology has made IoT devices more prone to severe security threats, such as Distributed Denial of Service (DDoS) attacks, which are dangerous threats to public systems and networks. Further, the frequency and complexity of IoT traffic-based DDoS attacks are increasing year-by-year. This article proposes an IoT traffic-based DDoS attack detection approach for classifying incoming IoT network traffic into 11 classes using multiclass machine learning techniques. The proposed approach comprises two phases: (i) designing and (ii) detection. In the designing phase, we employ the embedded feature reduction technique to create cost-effective and efficient classification models with a high feature reduction rate. Further, we evaluate these models using the K-fold cross-validation technique. While in the detection phase, we evaluate the performance of an efficient model by executing four different IoT traffic-based scenarios. A publicly available Bot-IoT dataset is employed to design and validate the proposed multiclass classification approach. The results show that the proposed approach provides an 84.4% feature reduction rate and approximately 5.19% higher classification accuracy than the existing approaches. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) paves the way for the modern smart industrial applications and cities. Trusted Authority acts as a sole control in monitoring and maintaining the communications between the IoT devices and the infrastructure. The communication between the IoT devices happens from one trusted entity of an area to the other by way of generating security certificates. Establishing trust by way of generating security certificates for the IoT devices in a smart city application can be of high cost and expensive. In order to facilitate this, a secure group authentication scheme that creates trust amongst a group of IoT devices owned by several entities has been proposed. The majority of proposed authentication techniques are made for individual device authentication and are also utilized for group authentication; nevertheless, a unique solution for group authentication is the Dickson polynomial based secure group authentication scheme. The secret keys used in our proposed authentication technique are generated using the Dickson polynomial, which enables the group to authenticate without generating an excessive amount of network traffic overhead. IoT devices' group authentication has made use of the Dickson polynomial. Blockchain technology is employed to enable secure, efficient, and fast data transfer among the unique IoT devices of each group deployed at different places. Also, the proposed secure group authentication scheme developed based on Dickson polynomials is resistant to replay, man-in-the-middle, tampering, side channel and signature forgeries, impersonation, and ephemeral key secret leakage attacks. In order to accomplish this, we have implemented a hardware-based physically unclonable function. Implementation has been carried using python language and deployed and tested on Blockchain using Ethereum Goerli's Testnet framework. Performance analysis has been carried out by choosing various benchmarks and found that the proposed framework outperforms its counterparts through various metrics. Different parameters are also utilized to assess the performance of the proposed blockchain framework and shows that it has better performance in terms of computation, communication, storage and latency. [ABSTRACT FROM AUTHOR]

P2P-based Edge Cloud (PEC) is widely used in Internet of Things (IoT). Inevitably, the sensor data routing technology has a significant impact on the performance of PEC. Due to its prevalence and complexity, the existing routing technologies in PEC need to be optimized. Specifically, key factors such as overall network traffic, user access latency, and resource utilization of edge nodes should be considered to adapt to the dynamic requirements of user request services and network topology. In order to address the challenges produced by these factors, an adaptive routing in P2P-based Edge Cloud is proposed, which is named ARPEC. In our approach, a target edge node selection scheme based on message activity and network topology is proposed, aiming to minimize the load on edge node and user access latency. Furthermore, to minimize system overhead, sensor data routing is mapped to minimum cost maximum flow (MCMF) graph. On this basis, a target edge node selection algorithm based on a grey linear regression combination prediction model is designed, and an incremental MCMF algorithm based on belief propagation (BP) is proposed. The evaluation results show that our approach can effectively improve PEC transmission performance and user experience. [ABSTRACT FROM AUTHOR]