Development of an optimized multimedia compression model for improved quality of service in virtual learning environments
Loading...
Date
2024-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Kampala International University
Abstract
In the rapidly evolving landscape of education, one of the most transformative advancements has been the advent of e-learning. This mode of education encompasses a wide range of activities, from online courses and virtual classrooms to interactive simulations and multimedia presentations, demanding substantial bandwidth. However, in low-developed countries, e-learning often relies on wireless and mobile networks with limited bandwidth, low computing devices, and poor network connectivity, thereby hindering the effectiveness of learning in virtual environments. This research aims to address this challenge by exploring data compression algorithms through a comprehensive comparative analysis and developing an optimized data compression model suitable for low-bandwidth networks. The goal is to achieve maximum reduction of data in transit, facilitating quicker and timelier transmissions, thereby enhancing virtual learning environments and ensuring an effective and improved learning experience in institutions of learning. The research objectives included establishing requirements for designing a multimedia compression model for optimal bandwidth utilization, developing a multimedia compression model with optimum bandwidth utilization, and evaluating the model through comparative performance analysis to attain more optimized results. This study employed the Design Science methodology, complemented by simulation and experimentation. Additionally, the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology was employed, involving problem understanding, data understanding, modeling, evaluation, and implementation phases. These methodologies guided the research process and facilitated the systematic development and assessment of the compression model. The approach allowed for concrete measurements and comparisons, providing robust evidence of the model's effectiveness. DCT, Huffman, and Fractural compression algorithms were studied, analyzed, and compared for use in the model design. All algorithms were applied to multimedia data, with a focus on images and videos. The best optimal results were obtained after combining DCT and Huffman, resulting in a compression ratio of 0.14745 compared to 0.148601 for DCT alone and 1.00330 for Fractural. Optimizing DCT with Huffman yielded superior compression ratio (CR) and peak signal-to-noise ratio (PSNR), making it best suited for use in low-computing devices and poor network environments, thereby providing optimal performance for improved quality of service in virtual learning environments.
Description
A thesis report submitted to the school of mathematics and computing in partial fulfilment of the requirement for the award of the degree of Master of Science in software engineering of Kampala International University