This paper proposes Snapping Shrimp Noise (SSN) detection and corrupted Orthogonal Frequency Division Multiplexing (OFDM) reconstruction methods to increase Bit Error Rate (BER) performance when OFDM transmitted signal is corrupted by impulsive SSNs in underwater acoustic communications. The proposed detection method utilizes multilayer wavelet packet decomposition for detecting impulsive and irregularly concentrated and SSN energy in specific frequency bands of SSN, and the proposed reconstruction scheme uses iterative decision directed-subcarrier reconstruction to recover corrupted OFDM signals using multiple carrier characteristics. Computer simulations were executed to show receiver operating characteristics curve for the detection performance and BER for the reconstruction. The practical ocean experiment of SAVEX 15 demonstrated that the proposed method exhibits a better detection performance compared with conventional detection method and improves BER by 250% and 1230% for uncoded and coded data, respectively, compared with the conventional reconstruction scheme.

提出了一种在水下声通信中，当冲激SSN破坏OFDM传输信号时，捕获虾噪声（SSN）检测和破坏的正交频分复用（OFDM）重建方法，以提高误码率（BER）性能的方法。所提出的检测方法利用多层小波包分解来检测SSN特定频带中的脉冲和不规则集中的SSN能量，并且所提出的重建方案使用迭代决策有向子载波重建来利用多载波特性来恢复损坏的OFDM信号。执行计算机仿真以显示接收器操作特性曲线（用于检测性能）和BER（用于重建）。