A Model-Driven Framework for Optimum Application Placement in Fog Computing Using a Machine Learning Based Approach

The invention of deep learning in secure routing triggered an exceptional expansion of the Internet of Things (IoTs). Concurrent analysis of human body data occurs in the fog layer after sensors and actuators collect information from smart medical devices. The combination of criticality with increased complexity and dynamic capabilities causes H-IoT devices to be incompatible with typical network configurations which generates security and QoS problems. The identification of appropriate fog nodes together with unnecessary data reduction proves to be a complicated process. This paper incorporates SDN-driven DL to design a secure and intelligent framework for H-CIoT networks which solves existing network challenges. The SDN architecture stands out as a suitable solution because it enables network infrastructure reconfiguration while managing distributed IoT network architecture through separate data and control planes. The Proposed ML based AODV and AOMDV offers enhanced network security through centralized control and programmability, allowing for fine-grained security policies and real-time adjustments. The AOMDV security module serves as an implementation to detect multiple attack types that appear in the IoT network. The training process of the Deep Learning model utilizes IOT devices archival data in industry. The system uses acquired information to determine if data needs to be transferred to the fog layer. The suggested framework utilizes deep learning hybridization and CNN for selecting the optimal fog node alongside its features. The simulation of the proposed framework demonstrated 99.59% accuracy and achieves 80% detection ratio together with a 0.99% ideal throughput and packet delivery rate of 0.89%, a minimum energy of 0.11 m joules, at a maximum speed of 0.84 bps, and a negligible delay of 0.3415 ms when tested with 30 nodes. alongside 4% increased F1-score performance at 10 ms faster latency and lower energy usage of 25 W and 0.66% better probability. Keywords:Machine Learning, ResNet, Deep Neural Network, CNN, Prediction Models, Hybrid Machine Learning, Routing Attacks Detection

Ahstract— With the rapid development of vehicular applications and mobile devices, demands for resources to process time-critical and computation-intensive tasks are increasingly prominent. In this paper, we propose a two-layer Vehicular Fog Computing (VFC) architecture, including the client layer and the fog layer. Vehicles may generate tasks as clients, which are further assigned to the nodes in the fog layer for processing. The fog layer aggregates available resources of vehicles and infrastructures by exploiting their communication, computation and storage capabilities. Each task requires certain amount of resources for processing at the fog nodes. We formulate a distributed task allocation (DTA) problem, which takes deadline, vehicle mobility and fog capacity into consideration, and aims at maximizing the overall resource utilization of system, via the cooperation of vehicles and fog nodes. We linearize DTA into a 0–1 integer linear programming (ILP) problem to obtain the optimal solution. Further, we design a heuristic algorithm to obtain near-optimal performance with low computational overhead, which decomposes DTA into two subprocess and schedules tasks in each fog node independently. Finally, we build the simulation model and conduct a series of experiments based on real-world vehicle trajectories, which demonstrate the effectiveness and scalability of the proposed algorithm.

Over the years, fog computing has emerged as a paradigm to complement the cloud computing in handling the delay sensitive IoT applications in a better manner. Using fog resources, better performance such as in-time service delivery, reduced network load, optimal energy usage etc. can be achieved. With such performance gain, users availing the IoT services are more satisfied. A well-known metric Quality of Experience (QoE), used to measure the satisfaction of IoT users, can be improved by enhancing the performance of the IoT applications. Fog computing is a geographically distributed paradigm and primary service of fog computing may not include the execution of offloaded tasks/applications from the IoT devices. This makes QoE aware placement of applications in fog computing a greater challenge. Since placement algorithm is itself a computational task and both IoT applications and fog nodes need a mediator fog node to execute the placement algorithm, the placement policy should be light weighted in terms of computational complexity. This work proposes a lightweight QoE aware application placement policy in fog computing using Modified TOPSIS that prioritizes the applications and fog instances based on their expectation and computational capability respectively for the placement. Modified TOPSIS inherits all the features of classical TOPSIS while it removes rank reversal problem of classical TOPSIS. Simulation experiments, for a comparative study, depict that the proposed model not only achieves the desired resource utilization, processing time, and reduced network congestion but reduces the application placement time also significantly compared to the state of art.

We proposed a network intrusion detection system that combines a stream machine learning model in the fog layer and an online labeling model in the cloud layer. The stream learning model is based on the Adaptive XGBoost machine learning algorithm, aiming to detect anomaly network traffic. The online labeling model is a batch machine learning model based on the Random Forest algorithm and is responsible to label unknown traffic and provide updates to the stream learning model in the fog layer. The proposed solution effectively detects abnormal traffic in the fog layer that is connected with IoT devices. The stream learning model updates the model at a lower cost as compared to the batch learning approach. To evaluate the proposed system, contemporary datasets are used to test the accuracy of the models. The experiment results show that the proposed scheme effectively achieves good classification accuracy with the cloud layer providing updates to the fog layer. The result is about 17.6% and 9.0% better than the baseline method for the UNSW-NB15 dataset and CIC-IDS2017 dataset, respectively. In addition, the stream learning approach can provide higher throughput than the batch learning approach.

IFogLearn++: A new platform for fog layer's IoT attack detection in critical infrastructure using machine learning and big data processing

A genetic algorithm for energy efficient fog layer resource management in context-aware smart cities

A communicable illness is one that spreads from person to person by a number of means, including contact with blood and body fluids, inhaling an airborne virus, or being bitten by an insect. Internet of Things (IoT)-oriented medical technologies have proliferated in prior decades, and many more applications depend on wearable sensors for versatility. Modern research has examined the use of software-defined networking (SDN)-oriented fog framework to create intelligent health care services. To choose the best fog nodes within the fog layers, however, to anticipate communicable disease detection.To solve this issue, the SDN-enabled Fog Node Selection for Early Detection of Communicable Diseases (FogCom) is created in this study. The FogCom picks the optimal fog nodes using game theory’s Subgame Perfect Nash Equilibrium (SPNE) method. The suggested method also takes into account the three-layered CNN methodology for detecting communicable illnesses. Efficiency analysis shows that the suggested FogCom outperforms earlier baseline techniques.

Fog computing (also known as edge computing) is a decentralized computing architecture that seeks to minimize service latency and average response time in IoT applications by providing compute and network services physically close to end-users. Fog environment consists of a network of fog nodes and IoT applications are composed of containerized microservices communicating with each other. Due to limited resources of fog nodes, it is often not possible to deploy all the containers of an application on a single fog node. Therefore, communicating containers need to be distributed on multiple fog nodes. Distribution and management of containerized IoT applications is always a critical issue to the system performance in a fog environment. Kubernetes, an open-source system, has grown into a container orchestration standard by simplifying the deployment and management of containerized applications. Despite the progress made by the academia and industry with respect to container management and the wide-scale acceptance of Kubernetes in cloud environments, container management in fog environment is still in the early stage in terms of research and practical deployment. This article aims to fill this gap by analyzing the expediency of Kubernetes container orchestration tool in the fog computing model. The paper also highlights limitations with the current Kubernetes approach and provide ideas for further research to adapt to the needs of the fog environment. Lastly, we provide experiments that demonstrate the feasibility and industrial practicality of deploying and managing containerized IoT applications in the fog computing environment.

With the emergence of Internet of Things (IoT) technology, a huge amount of data is generated, which is costly to transfer to the cloud data centers in terms of security, bandwidth, and latency. Fog computing is an efficient paradigm for locally processing and manipulating IoT-generated data. It is difficult to configure the fog nodes to provide all of the services required by the end devices because of the static configuration, poor processing, and storage capacities. To enhance fog nodes' capabilities, it is essential to reconfigure them to accommodate a broader range and variety of hosted services. In this study, we focus on the placement of fog services and their dynamic reconfiguration in response to the end-device requests. Due to its growing successes and popularity in the IoT era, the Decision Tree (DT) machine learning model is implemented to predict the occurrence of requests and events in advance. The DT model enables the fog nodes to predict requests for a specific service in advance and reconfigure the fog node accordingly. The performance of the proposed model is evaluated in terms of high throughput, minimized energy consumption, and dynamic fog node smart switching. The simulation results demonstrate a notable increase in the fog node hit ratios, scaling up to 99% for the majority of services concurrently with a substantial reduction in miss ratios. Furthermore, the energy consumption is greatly reduced by over 50% as compared to a static node.

Anomaly detection framework for Internet of things traffic using vector convolutional deep learning approach in fog environment

Low latency is critical to applications such as control of unmanned aerial vehicles. Such latency-sensitive services can be hosted closer to the user at the fog layer which can reduce overall latency through the reduction of transmission time and network congestion. To keep the latency low for mobile users connected to services deployed at the fog, these services need to be constantly migrated to follow the users. Unlike the cloud nodes, fog nodes are less reliable and are therefore subject to higher failure rate. In this paper, we propose an enhancement to the post-copy live migration algorithm to make it robust against failure. Simulation results show that robust migration reduces total migration time between 10-26% and downtime between 2-23% compared to non-robust migration. Furthermore, when the bandwidth to the backup node is lower, robust migration provides further improvement in both metrics.

This study addresses the interaction between IoT devices and their physical environment, highlighting its role in minimizing delays in IoT applications. The core challenge lies in scheduling IoT requests and efficiently allocating them to Fog and Cloud resources while ensuring fault tolerance. To tackle these issues, an intelligent optimization method based on the Plant Hormonal Defense Optimization (PDO) algorithm is proposed. The PDO algorithm, enhanced by applying penalties to restrict infeasible solutions, offers a high-quality solution within a reasonable computation time. The method's performance is evaluated using key metrics, such as average total delay in relation to data volumes, the number of lost requests, the delay difference between Fog and Cloud layers when altering processing speeds in the Fog layer, the 90th percentile delay as the number of servers in the Fog layer changes, and both Fog and Cloud breakpoints.The PDO algorithm is tested in a simulated environment that mimics real-world dynamics. Comparative analysis shows that the proposed method reduces delays by 23.84% to 48.51% when compared to other algorithms, demonstrating its superior performance.

In healthcare applications, numerous sensors and devices produce massive amounts of data which are the focus of critical tasks. Their management at the edge of the network can be done by Fog computing implementation. However, Fog Nodes suffer from lake of resources That could limit the time needed for final outcome/analytics. Fog Nodes could perform just a small number of tasks. A difficult decision concerns which tasks will perform locally by Fog Nodes. Each node should select such tasks carefully based on the current contextual information, for example, tasks’ priority, resource load, and resource availability. We suggest in this paper a Multi-Agent Fog Computing model for healthcare critical tasks management. The main role of the multi-agent system is mapping between three decision tables to optimize scheduling the critical tasks by assigning tasks with their priority, load in the network, and network resource availability. The first step is to decide whether a critical task can be processed locally; otherwise, the second step involves the sophisticated selection of the most suitable neighbor Fog Node to allocate it. If no Fog Node is capable of processing the task throughout the network, it is then sent to the Cloud facing the highest latency. We test the proposed scheme thoroughly, demonstrating its applicability and optimality at the edge of the network using iFogSim simulator and UTeM clinic data.

With the increasing popularity of Internet of Things (IoT) applications in many different areas, the round-trip delay of data processing in the cloud affect the user's quality perception. But fortunately, fog computing aims to service users at the network edge similar to cloud services, which helps in supporting IoT in processing the data near to the end-user, especially for time-sensitive applications. It makes the resource allocation of application placement requests in a fog environment more necessary to satisfy the Quality of Experience (QoE) Influence Factors (IFs). In this paper, an IoT application placement algorithm based on the Multi-Dimensional QoE (MD-QoE) model is proposed in a fog computing environment. The algorithm is composed of two main phases. The first phase is to prioritize different IoT application placement requests depending on three main domains of IFs which are: Environment runtime context, application usage, and user expectations considering the Quality of Service (QoS) violation as a feedback. The second phase is to map and place the request to the appropriate fog node instance, depending on its proximity, computing capabilities, and expected response time. The proposed algorithm is evaluated by simulating a fog environment using iFogSim. Experimental results indicate that the proposed algorithm significantly improves the QoE in respect of application placement time, application delay, network usage, and power consumption. Therefore, the proposed algorithm can improve the overall system performance with a slight increasing in power consumption in fog control nodes.

The Internet of Things (IoT) has boosted fog computing, which complements the cloud. This is critical for applications that need close user proximity. Efficient allocation of IoT applications to the fog, as well as fog device scheduling, enabling the realistic execution of IoT application deployment in the fog environment. The scheduling difficulties are multi-objective in nature, since they must handle the issues of avoiding resource waste, network latency, and maximising Quality of Service (QoS) on fog nodes. In this research, the Hybrid Multi-Objective Marine Predators Algorithm-based Clustering and Fog Picker (HMMPACFP) Technique is developed as a combinatorial model for tackling the problem of fog node allocation, with the goal of achieving dynamic scheduling using lightweight characteristics. Utilised Fog Picker to allocate IoT components to fog nodes based on QoS parameters. Simulation trials of the proposed HMMPACFP scheme utilising iMetal and iFogSim with Hypervolume (HV) and Generational Distance (IGD) demonstrated its superiority over the benchmarked methodologies utilised for evaluation. The combination of Fog Picker with the suggested HMMPACFP scheme resulted in 32.18% faster convergence, 26.92% more solution variety, and a better balance between exploration and exploitation rates.