JoWUA

Abstract

The rapid development of m-learning applications has facilitated collaborative e-learning in a large-scale environment in different geographical locations. Nevertheless, centralized approaches usually face scalability issues, network congestion, high latencies, and ineffective resource management when dealing with large numbers of users and highly volatile wireless networks. In order to solve this problem, this paper suggests using a Swarm Intelligence-Based Decentralized Mobile Learning Network (SI-DMLN), which helps increase the performance of collaborative e-learning in a large-scale network setting. The suggested model utilizes principles of swarm intelligence with an aim to make mobile learning nodes act autonomously through decentralization. The framework includes adaptive routing, distributed resource allocation, and decentralized node coordination in order to improve communication and learning processes independently from a central controller. Through interactions within the network, the network behavior changes and adapts to particular network loads and mobile characteristics. The results of experiments confirm improvements in network performance. A total throughput of 94.8 Mbps is obtained compared to centralized models (68.2 Mbps) and traditional routing algorithms (75.6 Mbps). There is a reduction in end-to-end latency to 85 ms compared to the previous models, showing over a 50% enhancement in performance compared to traditional centralized architectures. Moreover, there is an increase in Packet Delivery Ratio to 96.7% and a high reliability level of 0.96. It shows good fault tolerance in the system along with reliable communication between the nodes. Similarly, there is improved energy efficiency of the system by a value of 0.89 due to balanced consumption of the resources by all the mobile nodes. Comparing other performance metrics reveals improvements of 15-40% over their baselines. Hence, the findings support the conclusion that swarm intelligence-based optimization techniques yield efficient and scalable results for mobile e-learning networks. Thus, the suggested strategy can be applied in future collaborative e-learning frameworks based on wireless mobile technology. Possible research avenues include the application of predictive and adaptive learning models in future implementations.