Abstract
A wireless sensor network and a data sensing method thereof are provided. A prediction model is established according to the sensed data. When the statistical data, such as average data, of the sensed data is within a user allowable range, based on the prediction model, the average data of the sensed data is returned to the user, for reducing the sense times and the power consumption. Conversely, if the statistical data, such as average data, of the sensed data is beyond the user allowable range, based on the prediction model, the newly sensed data is real-timely returned to the user. Further, the prediction model is dynamically updated based on the subsequently sensed data.