The invention provides an adaptive interference cancellation method based on higher-order statistics for an L-DACS1 system. An interference scene of time-frequency domain overlapping between DME (Distance Measuring Equipment) and L-DACS1 is modeled into an interference quantitative model in which a deterministic signal is overlapped with a Gaussian colored noise, and a CE-LMS (Cumulant Error-Least Mean Square) adaptive filtering algorithm is provided accordingly. According to the algorithm, a third-order cumulant is taken as a higher-order statistic for reflecting a difference between the DME and the L-DACS1, a square of a third-order auto-cumulant of an error signal is taken as a cost function, and a variable-step mechanism based on a logarithmic spiral is introduced, so that the influence of the L-DACS1 on the DME is lowered, and meanwhile the DME interference resisting capability of the L-DACS1 is enhanced. Compared with an existing interference suppression scheme, the adaptive interference cancellation method has the advantages that a higher interference suppression ratio and a lower bit error rate can be achieved on the premise of keeping the complexity, convergence and accuracy constant, and a reference is provided for actual deployment of the L-DACS1 system.