Abstract: ‍ ‍Applications such as radar， sonar， and wireless communication have increased the requirements for the anti-interference ability and real-time performance of adaptive beamforming. The conventional adaptive beamforming algorithm based on the fastest iteration exhibits a sharp decline in interference suppression performance under low snapshot and coherent interference conditions and cannot effectively capture the signal characteristics to effectively suppress interference. Moreover， under interference disturbance and steering vector mismatch conditions， it cannot effectively suppress the interference. To address these issues， this study proposes a wide-null interference suppression adaptive beamforming method based on the steepest descent approach. The method first uses a spatial smoothing-based weight pre-processing mechanism to decorrelate signals， extract signal features， and construct a projection matrix to obtain initial weights. Then， it suppresses interference using the steepest descent algorithm to form the beam. Finally， a null broadening and deepening method， based on covariance matrix reconstruction， is used to further enhance the interference suppression capability and improve the algorithm’s robustness. Simulation results verified that the proposed algorithm effectively suppressed interference under wideband coherent interference conditions and demonstrated good robustness.