Table 1 Summary of related work.

¹⁷

NSL KDD

To develop a robust and efficient deep learning-based distributed attack detection framework for IoT networks

Feedforward neural networks and recurrent neural networks (RNNs)

Deploying and managing a distributed system across an extensive, diverse IoT network can be complex

¹⁸

NSL KDD

To develop a deep learning-based model for classifying multiple cyber-attacks, aiming to improve the accuracy and effectiveness of intrusion detection systems

Long-Short-Term Memory Recurrent Neural

Network (LSTM-RNN)

Complex Model

High False Alarm

Rate

¹⁹

NSL KDD

To investigate the use of a deep learning approach for detecting DDoS attacks, potentially improving the accuracy and effectiveness of DDoS detection methods

Deep Contractive Autoencoder (DCAE)

The availability and quality of the training data may limit the model’s performance, potentially leading to inaccurate or biased detection results

²⁰

NSL KDD

To enhance the speed of detection while upholding a commendable level of accuracy

SVM, logistic

regression,

KNN

The use of GPU technology results in decreased training and prediction time

²¹

NSL KDD

To categorize and predict various types of DDoSattacks through the application of machine learning

Random forest,

XGBoost

Improved accuracy may be achieved using an enhanced suggested model

²²

NSL KDD

To develop an optimized ensemble framework using big data analytics to effectively detect DDoS attacks targeting (IoT)

Convolutional

Neural Network (CNN) embedded with a Gated Recurrent Unit (GRU)

The model’s complexity results in high computational time

²³

NSL KDD

To develop a precise and effective DDoS attack detection system for IoT networks with a hybrid Sample Selected RNN-ELM model

Recurrent Neural Networks (RNNs) and Extreme Learning Machines (ELMs)

The model’s efficiency may hinge on the quality and variety of the training data and the particular attributes of the IoT network environment

²⁴

NSL KDD

To provide a resilient and privacy-conscious DDoS attack detection solution for diverse IoT contexts by integrating federated learning with explainable a50rtificial intelligence approaches

Explainable Artificial Intelligence (XAI) with Federated Deep Neural Networks (FDNNs)

The efficiency of this methodology may be affected by issues including communication latency, variability in device capabilities, and the intricacy of incorporating XAI algorithms into the federated learning framework

²⁵

NSL KDD

To develop a DDoS attack detection system using fine-tuned Multi-Layer Perceptrons

fine-tuned Multi-Layer Perceptron models

They are computationally intensive and slow

²⁶

NSL KDD

To develop a system by integrating machine learning techniques with an SDN controller framework

Support Vector Machines, Decision Trees

The system may need to be continuously updated and adapted to address new and evolving DDoS attack techniques effectively

²⁷

NSL KDD

To develop an intrusion detection system for IoT networks by integrating PCA for feature reduction with a CNN for accurate and efficient attack detection

Convolutional Neural Networks (CNN)

Computational overhead from CNN training could challenge resource-constrained IoT environments