A soil moisture classification model based on SVM used in agricultural WSN

The measurement of soil moisture is an important issue in the research and application of agricultural wireless sensor network. However, the measurement of soil moisture with high precision can hardly be implemented at a relatively low cost especially in a large scale. High cost and complexity of soil moisture sensor become the bottleneck of promotion of agricultural wireless sensor network. Actually, crops are favorable of the soil whose moisture lies in a certain range, which means the merely correct classification can meet the basic demands of regular agricultural applications. In this way, a soil moisture classification model based on Support Vector Machine (SVM) with low-cost sensors as a way of substitution for high-cost soil moisture sensors is proposed in this paper to meet the practical needs of agricultural wireless sensor network. Tests results show that the proposed model has good classification accuracy in the same soil environment.