Table 1 Summary of the reviewed literature.
References | Year | Research goal | Method | Limitation |
|---|---|---|---|---|
Bougueroua et al.14 | 2021 | Taxonomy for Collaborative Intrusion Detection Systems (CIDS) | Multi-Agent Systems, CIDS taxonomy | General overview, not specific implementation |
Alshahrani15 | 2021 | Detect malicious behavior in IoT devices | Machine learning (CoLL-IoT) | Performance dependent on UNSW-NB15 dataset characteristics |
Alsoufi et al.16 | 2021 | Analyze anomaly-based intrusion detection with deep learning in IoT | Systematic literature review, comparison of deep learning methods | Review-based, not a novel implementation |
Yadav et al.17 | 2022 | Design a deep learning-based IDS for IoT networks | Deep learning | Specific vulnerabilities to certain cybercrimes not fully explored |
Vishwakarma and Kesswani18 | 2022 | Develop a deep neural network-based IDS for IoT | Deep neural network, Netflow datasets, packet capture | High computational costs and packet inspection |
Saba et al.19 | 2022 | Develop an IDS based on CNN for IoT security threats | CNN | Potential for overfitting, performance dependent on IoT power |
Khan20 | 2021 | Develop a hybrid deep learning-based IDS | Convolutional Recurrent Neural Networks (CRNNs) | Computational complexity of CRNNs |
Luo21 | 2023 | Develop an IDS for distributed systems using SDN | Decision tree enhanced by Black-Hole based Optimization (BHO) | Performance dependent on network segmentation and BHO algorithm |
Otoum et al.22 | 2022 | Develop a Deep Learning-based IDS for IoT anomalies | Deep learning | General anomaly detection, specific attack type detection may vary |
Nasir et al.23 | 2022 | Detect intrusions in IoT traffic | DF-IDS model, SpiderMonkey, PCA, IG, CAE | Performance dependent on feature engineering and dataset characteristics |
Alani and Awad24 | 2022 | Develop a two-layer machine learning-based IDS for IoT | Machine learning, packet-based and flow-based classifiers | Performance dependent on feature extraction and dataset characteristics |
Ge et al.25 | 2021 | Develop an intrusion detection methodology for IoT | Deep learning, feed-forward neural network | Performance dependent on dataset and network model |
Ullah et al.26 | 2024 | Develop an enhanced IoT-IDS using multimodal big data and transfer learning | Multimodal big data representation, transfer learning | Vulnerability to flood attacks |
Choudhary et al.27 | 2024 | Develop an IDS for IoT environments using deep learning | CNN, Aquila optimization | Computational complexity of hybrid model |
Kaushik and Raweshidy28 | 2024 | Develop a Teaching–Learning-Based Optimization-enabled IDS (TLBO-IDS) | Teaching–Learning-Based Optimization (TLBO) | Performance dependent on TLBO algorithm and network characteristics |
Hanafi et al.29 | 2024 | Develop a HTTP-IDS for IoT using optimization and LSTM | Improved Binary Golden Jackal Optimization, Long Short-Term Memory (LSTM) | Performance dependent on NSL-KDD and CICIDS2017 datasets |
