Numerical Simulation and Performance Assessment of Improved Particle Swarm Optimization Based Request Scheduling in Edge Computing for IOT Applications

Main Article Content

Ravi Khandelwal, Manish Kumar Mukhija, Satish Kumar Alaria


Recent advancements in the cost, performance, and energy efficiency of IoT devices, network technologies (such as 5G), and distributed computing architectures have resulted in the explosive growth of the Internet and mobile connectivity, resulting in new distributed applications in areas like transportation, healthcare, mining, entertainment, and security, such as automated vehicles, augmented reality, and security. As a result, data has grown at an unprecedented rate, and the relevance of latency and regulation in data processing and management has increased. The new distributed applications may have bandwidth-hungry qualities (video surveillance, video conferencing, traffic monitoring), latency-critical characteristics (automated cars, robotic surgery, safety), and may produce activity spikes in specific locations or times (sporting events). High availability, minimal jitter, and security may be required by applications. Mobile edge computing (MEC) and ultradense computing (UDC) are becoming more crucial in the 5G future as the demands on compute and enormous data flows from the Internet of Things (IoT) increase. As a viable approach, task offloading provides Mobile users in the future will benefit from reduced latency and flexible processing.  UDEC is a network of universities and colleges. However, due to the restricted computing resources available at the university, It is made possible by edge clouds and the dynamic demands of mobile users.  It's difficult to assign compute demands to the right edge clouds. To accomplish this, we must first define the transmission power.  a problem of allocation (PA) for mobile users to save energy  consumption). Then we create a model of the joint problem.  as a mixed integer request offloading and resource scheduling (JRORS)  nonlinear programme to reduce the time it takes for a response  requests. The JRORS problem can be broken down into two parts.  Specifically, the request offloading (RO) and computing problems.  The challenge of resource scheduling (RS). As a result, we investigate as a dual decision-making dilemma, the JRORS problem and suggest a multi-objective optimization method based on MO-PSO, also known as improved particle swarm optimization. The outcomes of the simulation demonstrate that PSO can reduce transmission energy usage on the other hand, has a good convergence property.    In terms of reaction rate, it exceeds existing methodologies.  It can maintain a high level of performance in a fast-paced UDEC network.

Article Details