For the purpose of diagnosing aircraft health states effectively, a new method based on Empirical Mode Decomposition(EMD) combined with D-S evidence theory is proposed in this paper. First, original acoustic emission(AE) signals of aircraft structural components (stabilizer) were decomposed into several intrinsic mode functions(IMF) using EMD. Use the IMF to construct feature vectors of AE signal, and then apply Fuzzy Neural Network, GRNN and Elman neural network respectively to classify these vectors. Finally, D-S evidence theory is used for decision fusion to determine the aircraft health states. When compared with methods that use single classifier, the effectiveness of the proposed method is demonstrated by higher health diagnosis accuracy from experimental tests on certain type of aircraft.