Abstract: A self-estimation algorithm of residual energy is proposed, according to discharging process of battery and energy consumption model. The algorithm is improved to take into consideration the influence of different working conditions and the discharging time of battery. It is implemented and simulated on TinyOS, and also tested on real node device. Comparing the simulation data and measured data, the result indicates that the algorithm can more precisely predict work time and residual energy. For energy aware algorithm provide accurate energy basis, and improve the energy efficiency and life.