Aiming at the contradiction between detail information loss and noise reduction performance in aircraft telemetry vibration signal, a noise reduction method based on ensemble empirical mode decomposition (EEMD) and noise cancellation is proposed. The signal is processed by EEMD to obtain an intrinsic mode function (IMF), and the sum of the first order IMF component and the accumulated sum of other IMF components are respectively used as the reference noise and the signal to be denoised; The signal is mapped to a high-dimensional feature space by using a kernel method, and the reference noise and the signal to be denoised which are mapped to the high-dimensional feature space are used for noise cancellation. The computer simulation results show that the method has good noise reduction performance on the premise of avoiding the loss of signal detail information, and the processing results of the measured data in an aircraft test prove the effectiveness and practicability of the method.