Graph neural networks based queuing model for optimal load balancing in mobile ad hoc network

This paper proposes a new approach for optimizing traffic management in multiple access networks (MANETs) by utilizing the stream-enabled routing (SER) algorithm. The SER algorithm is used to determine which routing path is the most time- and resource-efficient. The proposed approach makes use of multipath routing in a manner that is consistent with the SER method. By combining the states of flows, queues, and links, a graph neural network (GNN)-based model attempts to break the circular dependencies that are described by these functions. The simulation is setup with joint parameters consisting of residual energy, packet delivery rate (PDR), and end-to-end delay. The results of the experiments show that the proposed protocol provides a significant improvement in terms of network efficiency when compared to using some baseline protocols designed for MANETs. A thorough framework for improving the performance of mobile ad-hoc networks (MANETs) is proposed in this paper. It is achieved by combining normal distribution-based sensor location with queueing theory-based graph neural network (QGNN) operations to maximize network efficiency, route selection, and congestion control. Through the use of sophisticated algorithms for congestion control and route optimization, the suggested technique creates MANET models with strategically placed sensors, which should greatly increase network capacity and dependability.image