Figure - available from: Soft Computing
This content is subject to copyright. Terms and conditions apply.
Source publication
The application of wireless sensor networks is not limited to a particular domain. Technology advancements result in innovative solutions for simple communication to large applications via wireless sensor IoT networks. Besides the advancements, there is a serious issue in terms of threats or attacks on wireless sensor networks, which is common. Var...
Similar publications
This paper presents flood prediction models for the state of Kerala in India by analyzing the monthly rainfall data and applying machine learning algorithms including Logistic Regression, K-Nearest Neighbors, Decision Trees, Random Forests, and Support Vector Machine. Although these models have shown high accuracy prediction of the occurrence of fl...
Citations
... Such approaches can dissect voluminous data about network traffic in search of very minute patterns indicative of wormhole attacks, which traditional rule-based systems would otherwise miss. In particular, machine learning models trained with supervised learning can be fed datasets containing normal and attack traffic to learn the features of wormhole attacks [13]. However, there exist challenges in implementing machine learning-based IDS in IoT networks. ...
... As a result, the load balance will increase in their research. This article [32] presented that feedforward, and fuzzy neural networks are used to create a unique IDS that can identify routing attacks in WSNs. Research results show that, in contrast to other methods such as support vector machine (SVM), decision tree (DT), and random forest (RF) designs, the suggested model achieves an average detection rate and the highest detection accuracy. ...
The Routing Protocol for Low-Power and Lossy Networks (RPL) routing protocol is utilized in the Internet of Everything (IoE) is highly vulnerable to various collaborative routing attacks. This attack can highly degrade network performance through increased delay, energy consumption, and unreliable data exchange. This critical vulnerability necessitates a robust intrusion detection system. This study aims to enhance a Collaborative Intrusion Detection System (CIDS) for detecting and mitigating joint attacks in the RPL protocol, focusing on improving detection accuracy while minimizing network delay and energy usage. A series of algorithms and techniques are implemented, including Queue and Workload-Aware RPL (QWL-RPL) for congestion reduction, weighted random forward RPL with a genetic algorithm for load balancing, fuzzy logic for trust evaluation, and Light Gradient Boosting Machine (GBM) for attack detection. Additionally, Q-learning with a trickle-time algorithm is used to classify and manage joint attacks effectively. Numerical analysis indicates that the proposed approach performs better than existing methods in multiple metrics, including accuracy, energy consumption, throughput, control message overhead, precision, and computing time. By integrating these diverse techniques, the proposed CIDS offers a scalable and efficient solution to improve the security and performance of RPL-based networks in IoE environments, outperforming current approaches in detection accuracy and resource optimization.
... Babouei 53 presented the adaptive neuro-fuzzy inference system (ANFIS) to identify patterns using control charts. By utilizing fuzzy and feed-forward neural networks, Ezhilarasi et al. 54 presented a unique intrusion detection method to identify routing attacks in wireless sensor networks. ...
Quality testing and monitoring advancements have allowed modern production processes to achieve extremely low failure rates, especially in the era of Industry 4.0. Such processes are known as high-yield processes, and their data set consists of an excess number of zeros. Count models such as Poisson, Negative Binomial (NB), and Conway-Maxwell-Poisson (COM-Poisson) are usually considered good candidates to model such data, but the excess zeros are larger than the number of zeros, which these models fit inherently. Hence, the zero-inflated version of these count models provides better fitness of high-quality data. Usually , linearly/non-linearly related variables are also associated with failure rate data; hence, regression models based on zero-inflated count models are used for model fitting. This study is designed to propose deep learning (DL) based control charts when the failure rate variables follow the zero-inflated COM-Poisson (ZICOM-Poisson) distribution because DL models can detect complicated non-linear patterns and relationships in data. Further, the proposed methods are compared with existing control charts based on neural networks, principal component analysis designed based on Poisson, NB, and zero-inflated Poisson (ZIP) and non-linear principal component analysis designed based on Poisson, NB, and ZIP. Using run length properties, the simulation study evaluates monitoring approaches, and a flight delay application illustrates the implementation of the research. The findings revealed that the proposed methods have outperformed all existing control charts.
... semantic-based [113][114][115], fuzzy logic [26,27,41,45,48,[116][117][118][119][120][121][122][123][124][125], game theory [126][127][128], bio-inspired [129,130] and Data Mining [42,131,132]. A graphical representation of the classification of AI techniques is shown in Figure 2.3. ...
Internet of Things (IoT) devices face unique security challenges due to their inherent limitations such as limited storage, low computational power, and energy-efficient wireless communication. Traditional security measures, designed for the legacy Internet, fail to adequately protect IoT devices and networks. Particularly vulnerable are Wireless Sensor Networks (WSN) and IoT networks that are susceptible to jamming—a type of attack that significantly threatens wireless networks due to their open nature and the simplicity of launching such attacks. Perpetrators can initiate jamming without specialized hardware or in-depth knowledge of the targeted system. Despite advances in wireless technologies, the ability to thwart jamming attacks in real-world scenarios remains limited, as evidenced by the vulnerability of current security protocols of cellular and Wi-Fi networks. This thesis addresses the critical need for practical anti-jamming strategies to enhance the security of wireless networks, particularly against intelligent jammers that employ advanced machine-learning algorithms to adapt to more sophisticated attack methods such as constant, deceptive, random, or reactive jamming. These intelligent attackers can adjust their strategies and even manipulate detection systems to evade identification. To counter these threats, this dissertation introduces a novel lightweight security framework that utilizes fuzzy logic algorithms to enhance the detection, localization, and recovery mechanisms against jamming attacks in IoT networks. The framework employs network layer metrics to detect jamming at the node level, utilizes a modified multilateration technique to accurately locate jammers, and implements recovery strategies by blacklisting the affected nodes and rerouting traffic within the RPL network. This thesis makes several noteworthy contributions representing a significant IoT security advancement. By applying fuzzy logic to combine crucial metrics from the data link and network layers, the proposed framework not only detects jamming incidents, but also precisely pinpoints their origin, which is essential for effective mitigation. This thesis performs accurate real-time detection and localization using data link and network-layer metrics collected and processed at the edge. Furthermore, the framework's capability to blacklist and recover from compromised network paths introduces a dynamic recovery mechanism that enhances network resilience. Additionally, this thesis introduces a novel jammer called the complex jammer, in which the proposed framework has been accurately identified. Moreover, the framework effectively demonstrates the suitability of fuzzy logic for accurately recognizing multiple jamming attacks in diverse situations, with high accuracy, low memory usage, and quick execution. The effectiveness of this framework was validated through extensive simulations, demonstrating its capability to handle multiple jammers and adapt to evolving jamming strategies, thus significantly improving the resilience of IoT networks against these pervasive threats.
... The training and testing results of the ANN demonstrate that this method can detect wormholes without the need for extra hardware and with a high detection accuracy of up to 97%. Fuzzy logic combined with a feed-forward neural network is a unique intrusion detection system that was presented by Ezhilarasi et al.'s research [52]. The neural network is trained using fuzzy rules, and simulation was performed to assess the neural network's performance. ...
A Mobile ad-hoc networks (MANET) is a wirelessly linked network of one or more devices that can configure itself. In a MANET, nodes can exchange data with one another directly or indirectly (via intermediary nodes). Because of the lack of central administration, open media, and several other reasons that make this type of network more vulnerable to security assaults, some researchers are utilizing artificial intelligence approaches in MANET routing to offer security. Several network layer attacks, including the black hole, and wormhole assaults, are covered in this essay. The detection of collaborative network assaults is examined, and frequent multiple network attacks are noted. A few of these assaults' symptoms will be emphasized. A network might exhibit a number of signs and observations that indicate the existence of an attack. The review calls for continued research to refine and deploy AI-based security mechanisms in real-world scenarios, addressing scalability concerns and advancing the vision of self-defending MANETs and wireless sensor networks (WSNs). The review serves as a resource for researchers, practitioners, and policymakers interested in fortifying the security of dynamic wireless networks.
... The application of the agricultural IoT integrates AI, IoT, blockchain, and virtual/augmented reality technologies. Ezhilarasi et al. (2023) investigate the combination of edge computing and AI, blockchain, and virtual/augmented reality technology. Intelligent video retrieval technology integrates video processing, computer vision, and AI, which greatly improves the efficiency of monitoring and the accuracy and linkage of monitoring systems. ...
At present, network attack means emerge in endlessly. The detection technology of network attack must be constantly updated and developed. Based on this, the two stages of network attack detection (feature selection and traffic classification) are discussed. The improved bat algorithm (O-BA) and the improved random forest algorithm (O-RF) are proposed for optimization. Moreover, the NIS system is designed based on the Agent concept. Finally, the simulation experiment is carried out on the real data platform. The results showed that the detection precision, accuracy, recall, and F1 score of O-BA are significantly higher than those of references [17], [18], [19], and [20], while the false positive rate is the opposite (P < 0.05). The detection precision, accuracy, recall, and F1 score of O-RF algorithm are significantly higher than those of Apriori, ID3, SVM, NSA, and O-RF algorithm, while the false positive rate is significantly lower than that of Apriori, ID3, SVM, NSA, and O-RF algorithm (P < 0.05).
... CatBoost offers a unique method of processing categorical data that involves little translation of categorical features. Changing from a non-numerical state to a numerical value can be a time-consuming and difficult task in feature engineering [15]. CatBoost inhibits this process, which also handles missing data and categorical data very well without encoding. ...
... A hybrid approach proposed in [32] combined the definition of fuzzy logic with the learning ability of neural networks to detect routing attacks in WSNs. The success of this solution depends on the quality and quantity of training data, the creation of fuzzy rules, the neural network architecture and the starting point. ...
... Performance measures such as detection rate (DR), false-positive rate (FPR), accuracy, and precision given in Table 6 are derived from the confusion-matrix values and examined for various attack scenarios. These matrices have the following mathematical formulation [32]: The number of correctly identified attacks divided by the total number of attacks yields the detection rate (DR). It is extremely rare for a classifier to obtain a TPR of 1, which indicates that every intrusion is successfully recognized. ...
... Feature selection results in the fewest incorrect classifications, yet the proposed system performs better than other detection methods, as shown by the analysis in Figure 14. This analysis compares the detection rate of different routing attacks by the proposed system with that of a technique proposed in [32] and those of machine-learning techniques like support-vector machines (SVM), DT, and RF. It is observed from the results that the maximum detection performance is attained by the proposed approach. ...
Wireless sensor networks (WSNs) are essential in many areas, from healthcare to environmental monitoring. However, WSNs are vulnerable to routing attacks that might jeopardize network performance and data integrity due to their inherent vulnerabilities. This work suggests a unique method for enhancing WSN security through the detection of routing threats using feed-forward artificial neural networks (ANNs). The proposed solution makes use of ANNs’ learning capabilities to model the network’s dynamic behavior and recognize routing attacks like black-hole, gray-hole, and wormhole attacks. CICIDS2017 is a heterogeneous dataset that was used to train and test the proposed system in order to guarantee its robustness and adaptability. The system’s ability to recognize both known and novel attack patterns enhances its efficacy in real-world deployment. Experimental assessments using an NS2 simulator show how well the proposed method works to improve routing protocol security. The proposed system’s performance was assessed using a confusion matrix. The simulation and analysis demonstrated how much better the proposed system performs compared to the existing methods for routing attack detection. With an average detection rate of 99.21% and a high accuracy of 99.49%, the proposed system minimizes the rate of false positives. The study advances secure communication in WSNs and provides a reliable means of protecting sensitive data in resource-constrained settings.
... Intrusions are typically described as malicious actions aimed at acquiring network entry and executing unauthorized operations. The network is secured by detecting any unauthorized and malicious activities through the use of an IDS [13,14]. ...
A wireless Sensor Network (WSN) is made up of many sensor nodes which gather and transmit data to a central location. The limited resources of the nodes create significant security challenges when deploying and communicating WSNs. The detection of unauthorized access is a crucial aspect of enhancing the security measures of WSNs. The utilization of network intrusion detection systems (IDS) has become an essential aspect of any communication network, as they offer valuable services to the network. Several studies in the field of machine learning have been conducted to explore the potential of utilizing this technology for intrusion detection in WSNs, yielding promising outcomes. These efforts still need to be more precise and efficient against network traffic unbalanced data issues. The paper presents a new model for detecting intrusion attacks that utilize a hybrid multilayer perceptron (MLP) and CatBoost classifier, as well as feature selection techniques. The proposed approach aims for good performance in identifying different forms of threats. The system performs data preprocessing on various datasets and reduces the dataset size using a feature selection algorithm. Pelican Optimization Algorithm (POA) has been proposed for tuning the hyper-parameters of the classifier designs and selecting the relevant features from the dataset. The CSE-CIC-IDS2018, AWID, and UNSW-NB15 databases reutilized for conducting performance evaluations on the proposed framework. The tests included accuracy, precision, recall, FAR, DR and complexity time. The proposed model has a low FPR and high accuracy in binary classification, as shown.
... Many techniques and methods have been developed to deal with these attacks. For example, various intrusion detection methods have been developed to detect common network attacks, focusing on other routing attacks such as black hole attacks, Sybil attacks, identity replication attacks, selective forwarding attacks, wormhole attacks, and hello flood attacks [12]. Prasad et al. investigated the detection method that can classify benign and malicious information in the MANET networks based only on routing attacks [13]. ...
... Ezhilarasi et al. Introduced in 2022, a new intrusion detection system that uses fuzzy and feed-forward neural networks to detect only routing attacks in wireless sensor networks [12]. The two previously cited papers work on routing attacks that include a set of attacks according to [12]. ...
... Introduced in 2022, a new intrusion detection system that uses fuzzy and feed-forward neural networks to detect only routing attacks in wireless sensor networks [12]. The two previously cited papers work on routing attacks that include a set of attacks according to [12]. In this paper, our work is based on Black Hole and Wormhole attacks because they are the major attacks in a MANET [14][15][16]. ...
This paper addresses the security concerns associated with Mobile Ad-hoc Networks (MANET) and proposes a new method for detecting and preventing attacks using machine learning. The study involved the creation of a MANET with 26 nodes in NetSim (Network Simulator) software, followed by the implementation of wormhole and blackhole attacks. A dataset was generated from the network traffic obtained during the simulations, and a machine-learning model was designed to predict and detect these attacks. The model achieved high sensitivity, accuracy and f1 scores of 99%. The effectiveness of the model was tested by developing a real-time application. This method can be applied to any wireless network and is particularly relevant for companies that use Ad-hoc networks for communication.
