Abstract

This study examines edge computing architectures integrated with Internet of Things (IoT) sensor networks for real-time agricultural decision support. By synthesizing recent literature (2014–2023), the paper evaluates architectural models, performance trade-offs, and operational challenges associated with deploying edge-enabled smart agriculture systems. Quantitative evidence from prior studies indicates that edge-based processing can reduce end-to-end latency by 40–65%, lower network bandwidth consumption by 30–70% and improve energy efficiency of sensor networks by up to 45% compared to cloud-centric architectures. The analysis further highlights improvements in real-time irrigation control, pest detection accuracy, and fault tolerance under intermittent connectivity. Key challenges related to interoperability, security, scalability, and cost are critically assessed. The findings underscore edge computing as a foundational enabler for autonomous, resilient, and data-driven agricultural systems, while identifying research gaps for standardized benchmarks, large-scale field validation, and energy-aware edge intelligence.