An Empirical Model for 2.4 GHz Radio Propagation in a Gala Apple Orchard and Evaluation of the Model Performance by Simulation

Abstract

<abstract><title><italic>Abstract. </italic></title> Apple production is high in China, but problems such as extensive management and poor technology are common in the Chinese apple industry. A wireless sensor network (WSN) can be used to monitor environmental information to enable timely and appropriate decision-making on pest prevention or to provide the necessary nutrition based on the collected data, thus improving apple quality and yield. The WSN nodes communicate with each other via wireless signals; therefore, a propagation model is needed to deploy the nodes and facilitate the application of a WSN in an apple orchard. Most of the existing empirical propagation models do not include environmental parameters; however, the density of leaves and the size of apples vary with time in an apple orchard from spring to winter. A 2.4 GHz empirical propagation model was developed for a gala apple orchard, including two parameters related to the environment: the month and the antenna height. Field tests were performed on the same line of apple trees for different months and at different heights in a gala apple orchard in China’s Shandong province. The tests were repeated from March to December in the middle of each month. The path losses at 18 distance points and six heights were collected in each test. An empirical model was created using the month, the antenna height, and the propagation distance as parameters to estimate the path loss in the horizontal direction along a line of gala apple trees. Validation experiments were also performed along a line of trees at five different heights from March to December in a different gala apple orchard. The validation results showed that the new model could estimate the path loss in the gala apple orchard. Most of the determination coefficients were larger than 0.7, and most of root mean square errors (RMSE) were less than 5. Simulation experiments were also conducted to evaluate the performance of the new model in applying a WSN to a gala apple orchard.