Volume 14, Issue 2, June 2023

A Comparative Study of GA and PSO Algorithm in Cloud Computing and IoT Technology
Pages: 1-8 (8) | [Full Text] PDF (379K)
Sharjeel Tariq, Muhammad Shahzad Ashraf Rana, M. Junaid Arshad
Department of Computer Science, University of Eingineering and Technology, Lahore-Pakistan

Abstract -
Web administration organizations are excellent at organizing creative applications for various Internet-based business arrangements. This paper uses two metaheuristic calculations, specifically Genetic Algorithm (GA) and Particle Swarm Optimization Algorithm (PSO), to handle QoS-based assistance organization issues. Quality of service has transformed into a fundamental issue in the administration of web administrations, given the significant number of administrations that outfit comparative usefulness yet with different qualities. This paper compares two popular optimization algorithms: Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). To assess the performance of the two algorithms, they are tested on two well-known benchmark functions and two engineering optimization problems, namely, the welding sequence problem and the design optimization of a truss structure. The results indicate that both algorithms can find the optimal global solutions for the benchmark functions, but PSO shows a better accuracy and faster convergence speed. However, when applied to engineering optimization problems, the superiority of PSO is not as evident. GA demonstrates better performance in finding the optimal solution with fewer evaluations. The authors conclude that selecting the appropriate algorithm for solving an optimization problem depends on the specific characteristics of the problem, such as the number of variables, the complexity of the objective function, and the constraints.
 Index Terms - Genetic Algorithm, Particle Swarm Optimization Algorithm, Workflow, Saas, Iaas, Paas, Quality of Service, IoT, Service Composition and Workflow Applications
An Enhanced Behavioral Modelling Approach using Blockchain Technology: A brief Survey
Pages: 9-16 (8) | [Full Text] PDF (324K)
Lea Yue, Kung Yin
Department of Computer Science and Information Technology, Korea

Abstract -
Since the organizations have become the large enough and more complex with the passage of time, thus the organizations demand for a mechanism helping them in unfolding their systems business processes in such a way that it can be easily presentable and understandable by the software experts as well as for the business professionals, also that they can easily communicate with each other to fulfill their demands and needs. This study has highlighted some gaps of activity diagram in context of modeling business process using unified modeling language such as process of data flow, messaging, transaction and multiple instances etc. and suggested some ways to overcome the highlighted issues and limitations by importing facilities from business process theory into unified modeling language.
 Index Terms - Business Process Model, Unified Modeling Language (UML), Technique, Emotional and Behavioral Modelling
Comparative Analysis of Machine Learning Algorithms for Intrusion Detection in IoT Environment
Pages: 17-24 (8) | [Full Text] PDF (465K)
Hafsa Riaz, Anum Aslam, M. J. Arshad
Department of Computer Science, University of Engineering and Technology, Lahore, Pakistan

Abstract -
The cumulative number of connected devices in IoT environments has led to a corresponding increase in the number of security threats. For IoT networks and devices to be secure and private, intrusion detection is essential. In the Internet of Things, machine learning algorithms have become a promising intrusion detection method. However, there are several different machine learning algorithms to choose from, each with its own strengths and weaknesses. A comparison of frequently employed machine learning techniques for intrusion detection in IoT contexts is presented in this review study. The paper examines the strengths and weaknesses of each algorithm, including their ability to detect known and unknown attacks, their false positive rates, their computational efficiency, and their training data requirements. Several machine learning algorithms, including Support Vector Machines, Artificial Neural Networks (ANN), Logistic Regression (LR), Decision Trees (DT), K-Nearest Neighbour (kNN), Random Forest (RF), Naive Bayes, and Deep Learning, are examined in-depth in this paper. The analysis includes a discussion of the algorithms' performance in different use cases, as well as their potential limitations. The paper concludes with recommendations for selecting the best machine learning algorithm for intrusion detection in IoT environments. The recommendations consider the specific use case, available data, and other relevant factors. The paper provides valuable insights for organizations looking to improve their IoT security posture and protect their devices and networks from potential threats.
 Index Terms - Machine Learning Algorithms, Deep Learning, Iot Environment, Intrusion Detection, K-Nearest Neighbors and Logistic Regression
NodeMCU and Cloud Computing for IoT: A Review of Integration Strategies
Pages: 25-29 (5) | [Full Text] PDF (234K)
Zuhaib B., Saad Bin Ghias, Muizz B., M. J. Arshad
Department of Computer Science, University of Engineering and Technology, Lahore, Pakistan

Abstract -
This review paper explores the integration of NodeMCU, an open-source IoT development board, with cloud computing platforms such as AWS and Azure. The paper begins by providing an introduction to NodeMCU and cloud computing, followed by a discussion of the benefits of integrating these technologies. It then reviews the various communication protocols and APIs used for integration, and compares the features and capabilities of different cloud platforms. The paper also examines scalability and security considerations, and offers best practices for securing NodeMCU-based IoT systems in the cloud. Finally, the paper discusses the limitations and challenges of integration, emerging trends and technologies, and future research directions. Overall, this review paper serves as a comprehensive guide for developers and researchers interested in exploring the integration of NodeMCU with cloud computing platforms for building scalable and secure IoT applications.
 Index Terms - NodeMCU, IoT, Cloud Computing, AWS, Azure, Communication Protocols, APIs, Scalability, Security, Best Practices, Emerging Trends and Research Directions