Date of Publication :7th December 2025

Abstract: Wireless Sensor Networks (WSNs) are critical for modern applications such as environmental monitoring, industrial automation, and smart cities. Traditional network topologies like mesh and torus offer reliability and redundancy but face limitations including high energy consumption, edge congestion, and rigid structure, especially in large-scale deployments. Gaussian connection models address these challenges by adopting a probabilistic and adaptive clustering approach. Nodes connect based on a Gaussian distribution considering distance, energy level, and other metrics, enabling balanced load distribution, energy efficiency, and robust fault tolerance. Advanced algorithms such as LEGN and TEGN leverage Gaussian models to optimize cluster head selection and routing, significantly improving network lifetime and reducing packet loss and latency. Comparative analysis shows Gaussian models outperform mesh and torus topologies in scalability, energy efficiency, and adaptability. This paper highlights their advantages through algorithmic insights, implementation case studies, and performance comparisons, justifying Gaussian connection models as an optimal choice for next-generation WSNs. Future work will explore hybrid models and deeper reinforcement learning to further enhance performance.

Reference :