Abstract: In order to avoid the whole wireless sensor networks (WSN) dead because of the high price of the measure precision, time-synchronization, hardware and power, so as to improve the precision of localization, an improved DV-Hop localization algorithm, which is simpler and better for the typical localization algorithm, is proposed based on the reality environment of breeding field. And its capability of localization error was evaluated by simulate and experiment. The results show that, the localization error of the improved algorithm is 3.17% less than that of the formerly algorithm, can measure the track of the moving sensor nodes as animals, and its average error is 0.33 m between reality position and the tested position. The improved algorithm is not only without any assistance of hardware such as GPS, but also cuts down the price of calculation, economizes the energy of sensor nodes, improves the precision of localization, applies available in the reality environment.