There are requirements of low power consumption and low cost for terminal nodes in agricultural wireless sensor networks（ WSNs）

however

the existing soil moisture sensors are difficult to meet the requirements. A method to measure soil moisture by fusing datas of soil resistance sensor with soil temperature sensor is proposed

meanwhile

multiple linear regression and BP neural network are used to set up measurement model of soil moisture. The method is applied in agricultural WSNs based on Zig Bee. The results of experiments show that precision of the proposed method can reach ± 2. 5 % RH in the same type of soil

which means that the proposed method can be used for practical soil moisture measurement in agricultural field.