Fly: Femtolet-based edge-cloud framework for crop yield prediction using bidirectional long short-term memory

Abstract

Crop yield prediction is a crucial area in agriculture that has a large impact on the economy of a country. This article proposes a crop yield prediction framework based on Internet of Things and edge computing. We have used a fifth generation network device referred to as femtolet as the edge device. The femtolet is a small cell base station that has high storage and high processing ability. The sensor nodes collect the soil and environmental data, and then the collected data is sent to the femtolet through the microcontrollers. The femtolet retrieves the weather-related data from the cloud, and then processes the sensor data and weather-related data using Bi-LSTM. The femtolet after processing the data sends the generated results to the cloud. The user can access the results from the cloud to predict the suitable crop for his/her land. This is observed that the suggested framework provides better accuracy, precision, recall, and F1-score compared to the state-of-the-art crop yield prediction frameworks. This is also demonstrated that the use of femtolet reduces the latency by ˜25% than the conventional edge-cloud framework.