Abstract: ‍ ‍L-band digital aviation communication system （L-DACS） is the future air-ground data link for the route stage. The working frequency band of L-DACS is set up between two channels of distance measure equipment （DME）. In order to eliminate the interference of the high-power pulse signal generated by the DME on the orthogonal frequency-division multiplexing receiver of the L-DACS forward link， this article proposes a high-power DME pulse interference signal suppression method based on extended block sparse Bayesian learning-boundary optimization （EBSBL-BO） algorithm. Firstly， the compressed sensing model of the DME interference signal is constructed using the empty sub-carriers signals of the orthogonal frequency division multiplexing receiver of the L-DACS. Then the high-power pulse signal recoverd according to the EBSBL-BO algorithm. Finally， the reconstructed high-power DME pulse signals are removed in the time domain. Compared with other sparse Bayesian reconstruction algorithms， the algorithm proposed in this paper has higher reconstruction accuracy of DME pulse signal and lower bit error rate of OFDM receiver， which can further reduce bit error rate of orthogonal frequency-division multiplexing receiver and enhance the receiving performance of the L-DACS effectively.