Abstract: Abstract: In order to realize the real time remote monitoring and precise management of the environment in litchi orchard, an information acquisition and intelligent irrigation expert decision system based on Internet of things in litchi orchard was designed. The system collected real-time environmental information such as soil moisture, air temperature and humidity, light intensity, wind speed and rainfall of litchi orchard through information collecting terminal modules then transmitted data packets of environmental information to gateway through wireless sensor network. The data packets would be processed and then transmitted to the cloud server through the GPRS network by the gateway. Combined with the environmental data and expert knowledge, the expert-decision system would establish multiple mathematical models of decision to achieve some decision-making functions such as the crop water requirements computing, irrigation time prediction, optimum quantitative irrigation decision-making and irrigation scheduling. The decision results were fed back to the control terminal module for intelligent controlling. The system was tested at the litchi orchard of College of Horticulture in South China Agricultural University from Feb. 11th 2016 to Mar. 15th 2016. In the test of the monitoring accuracy of information collecting terminal module, the results obtained from TES-1317 resistance thermometer, DT-8896 wet and dry bulb hygrometer, AR813A light illumination measuring instrument and so on were compared with those obtained information collecting terminal module. The results revealed the maximum relative error of the temperature 2.98%, that of the humidity 3.06%, that of the illumination was 6.25%, that of soil moisture 4.13%, that of wind speed 4.76% and precipitation 2.25%. It indicated that the monitoring accuracy of information collecting terminal module was high and was able to provide precise decision-making data for expert decision system. The decision-making system with multi-parameter and single parameter was tested to measure the accuracy of the expert decision system and determine the accuracy of the system according to the comparison between drought condition and actual situation. The result showed that the accuracy of multi-parameter decision was higher than the simple single parameter decision. The system gained the number of days automatically during the tests of the decision-making instantaneity of the system. When the soil water content was lower than the minimum of the litchi optimum water content which was 15.55%, the system opened the solenoid valve and began to do drip irrigation. On the other hand, the system stopped doing drip irrigation when the soil water content was higher than 19.14%. The result showed that the average soil water content of irrigation area was 17.4%, satisfying the requirement of the soil water content in the growth process of litchi trees, which illustrated that the irrigation decision made by system had strong real-time. The result from subtracting the days interval for predicting irrigation of the system by the days when the soil water content was below the minimum from the system began to irrigate showed that the system prediction accuracy could reach 75%, which illustrated that the predicting real-time of the system was better. This system realizes environmental information acquisition and intelligent irrigation in litchi orchard, guiding the users manage the litchi orchard better.