Optimizing IoT Data Transmission in Smart Agriculture using Cloud Computing

This research presents a novel flow meter reading identification technique utilizing linear regression optimization, tailored for agricultural applications and optimized for edge computing environments. The study involves creating a diverse dataset of flow meter readings from agricultural settings and preprocessing the data to address anomalies. Leveraging linear regression as the fundamental model, the algorithm undergoes training and evaluation phases, considering metrics such as Mean Squared Error (MSE). While recent advancements in edge computing for agriculture encompass CNNs, RNNs, and hybrid methods, challenges persist, including limited resources on edge devices, real-time processing requirements, environmental variability, and robustness across diverse agricultural scenarios. The proposed method is refined to optimize efficiency and resource utilization, presenting a lightweight architecture suitable for deployment on edge devices within agricultural environment, aligning with the constraints of edge computing infrastructure.