An Amendable Multi-Function Control Method using Federated Learning for Smart Sensors in Agricultural Production Improvements

Abstract

Communications and Computer Engineering Department, Faculty of Engineering, Al-Ahliyya Amman University, Amman 19328, Jordan School of Information Technology, Skyline University, Sharjah, 1797, UAE Smart Sensors are used for monitoring, sensing, and actuating controls in small and large-scale agricultural plots. From soil features to crop health and climatic observations, the smart sensors integrate with sophisticated technologies such as the Internet of Things or cloud for decentralized processing and global actuation. Considering this integration, an Amendable Multi-Function Sensor Control (AMFSC) is introduced in this proposal. This proposed method focuses on sensor operations that aid agricultural production improvements. The agriculture hindering features from the soil, temperature, and crop infections are sensed and response is actuated based on controlled operations. The control operations are performed according to the sensor control validation and modified control acute sensor, which helps to maximize productivity. The sensor control and operations are determined using federated learning from the accumulated data in the previous sensing intervals. This learning validates the current sensor data with the optimal data stored for different crops and environmental factors in the past. Depending on the computed, sensed, and optimal (adaptable) data, the sensor operation for actuation is modified. This modification is recommended for crop and agriculture development to maximize agricultural productivity. In particular, the sensing and actuation operations of the smart sensors for different intervals are modified to maximize production and adaptability. The efficiency of the system was evaluated using different parameters and the system maximizes the analysis rate (12.52%), control rate (7%), adaptability (9.65%) and minimizes the analysis time (7.12%), and actuation lag (8.97%)