Enhancing Smart Farming with IoT Sensors Using FRPGW and HALSTM for Accurate Predictions

This Internet of Things (IoT)-based real-time data collection and analysis system enhances the productivity of agriculture. The use of IoT sensors in monitoring soil conditions optimizes the agricultural methods to resolve problems such as wasteful resource consumption and high operating costs resulting from the lack of accurate, current data and the manual interventions made in the entire process. These data are subjected to pre-processing, including normalization, which normalizes the data scale, and noise filtering to eliminate inaccuracies. Statistical measures are used to calculate the mean, median, skewness, and kurtosis of the data. Feature extraction is applied to derive meaningful insights from the data. Fused Red Piranha Grey Wolf Optimization (FRPGW) algorithm determines the relevant features that can be applied to the accurate models. Crop productivity and drought conditions are predicted by the Hybrid Artificial Long Short-Term Memory (HALSTM) model. It improves resource management, decision-making, and productivity in farms