The domain of Internet of Things (IoT) has witnessed immense adaptability over the last few years by drastically transforming human lives to automate their ordinary daily tasks. This is achieved by interconnecting heterogeneous physical devices with different functionalities. Consequently, the rate of cyber threats has also been raised with the expansion of IoT networks which puts data integrity and stability on stake. In order to secure data from misuse and unusual attempts, several intrusion detection systems (IDSs) have been proposed to detect the malicious activities on the basis of predefined attack patterns. The rapid increase in such kind of attacks requires improvements in the existing IDS. Machine learning has become the key solution to improve intrusion detection systems. In this study, an ensemble-based intrusion detection model has been proposed. In the proposed model, logistic regression, naive Bayes, and decision tree have been deployed with voting classifier after analyzing model’s performance with some prominent existing state-of-the-art techniques. Moreover, the effectiveness of the proposed model has been analyzed using CICIDS2017 dataset. The results illustrate significant improvement in terms of accuracy as compared to existing models in terms of both binary and multi-class classification scenarios.

A large number of smart devices in Internet of Things (IoT) environments communicate via different messaging protocols. Message Queuing Telemetry Transport (MQTT) is a widely used publish–subscribe-based protocol for the communication of sensor or event data. The publish–subscribe strategy makes it more attractive for intruders and thus increases the number of possible attacks over MQTT. In this paper, we proposed a Deep Neural Network (DNN) for intrusion detection in the MQTT-based protocol and also compared its performance with other traditional machine learning (ML) algorithms, such as a Naive Bayes (NB), Random Forest (RF), k-Nearest Neighbour (kNN), Decision Tree (DT), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRUs). The performance is proved using two different publicly available datasets, including (1) MQTT-IoT-IDS2020 and (2) a dataset with three different types of attacks, such as Man in the Middle (MitM), Intrusion in the network, and Denial of Services (DoS). The MQTT-IoT-IDS2020 contains three abstract-level features, including Uni-Flow, Bi-Flow, and Packet-Flow. The results for the first dataset and binary classification show that the DNN-based model achieved 99.92%, 99.75%, and 94.94% accuracies for Uni-flow, Bi-flow, and Packet-flow, respectively. However, in the case of multi-label classification, these accuracies reduced to 97.08%, 98.12%, and 90.79%, respectively. On the other hand, the proposed DNN model attains the highest accuracy of 97.13% against LSTM and GRUs for the second dataset.

With the increased penetration of cloud computing and virtualization, a plethora of internet of things devices have been deployed globally. As a result, computationally intensive tasks are transmitted from the edge towards the centralized cloud for processing that leads to increased energy utilization in the cloud data centers while at the same increasing significant latency for critical applications. Recent years have witnessed a paradigm shift from centralized cloud computing towards mobile edge computing (MEC), where computational tasks are offloaded at the edge servers near user equipment (UE). This paradigm leads to lowering the energy utilization in the cloud data centers, along with low latency for UE and efficient resource utilization at the edge. In this context, the scale and complexity of the MEC networks is drastically increasing and, consequently, finding effective energy-efficient solutions for computational task offloading and resource allocation in MEC networks has become an ambitious task. To address the aforementioned challenges, this work surveys the state of the art in different categorizations of algorithm-based computational task offloading and resource allocation strategies focusing on energy utilization. It also provides a detailed cross-comparison of existing strategies in terms of their implementation specifications. Additionally, this paper also highlights open challenges and potential future research directions to facilitate efficient task offloading and resource allocation at the edge with reduced energy consumption at the centralized data centers. Our work also paves the way for the deployment of critical applications at the edge that require low latency and high service quality guarantees.

The widespread adoption of Autonomous Vehicles (AVs) necessitates Mobile Edge Computing (MEC) to offload computation-intensive tasks, thereby minimising energy consumption while maintaining real-time responsiveness for critical operations such as obstacle avoidance, lane keeping, adaptive cruise control, and intersection management. However, optimising task offloading and resource allocation poses significant challenges due to the AVs' constrained battery life and limited processing capacity. To tackle this, we propose a unified framework integrating Optimal Stopping Theory (OST), Deep Reinforcement Learning (DDPG), and a Genetic Algorithm (GA). OST determines the optimal offloading checkpoint, while DDPG dynamically optimises resource allocation to minimise energy consumption and ensure rapid responsiveness for realtime autonomous operations. Additionally, GA enhances MEC efficiency by optimising the execution order in scenarios where multiple tasks need to be offloaded. Simulation results show that OST improves offloading performance by at least 9.4% compared to conventional strategies. The DDPG-based approach reduces vehicle energy consumption by at least 24.7%, while the GA method decreases queue waiting times by 67.6% and reduces failed offloading tasks by 20%. This framework effectively balances energy efficiency and task latency while ensuring reliable and responsive AV operations in the MEC-assisted vehicular network.