Underwater acoustic recognition for ship-radiated signals has high practical application value due to the ability to recognize non-line-of-sight targets. However, due to the difficulty of data acquisition, the collected signals are scarce in quantity and mainly composed of mechanical periodic noise. According to the experiments, we observe that the repeatability of periodic signals leads to a double-descent phenomenon, which indicates a significant local bias toward repeated samples. To address this issue, we propose a strategy based on cross-entropy to prune excessively similar segments in training data. Furthermore, to compensate for the reduction of training data, we generate noisy samples and apply smoothness-inducing regularization based on KL divergence to mitigate overfitting. Experiments show that our proposed data pruning and regularization strategy can bring stable benefits and our framework significantly outperforms the state-of-the-art in low-resource scenarios.

翻译：针对舰船辐射信号的水声识别具有极高的实际应用价值，因其能够实现对非视距目标的识别。然而，受限于数据采集难度，获取的信号不仅数量稀缺，且主要由机械周期性噪声构成。实验表明，周期性信号的重复特性会导致双下降现象，即模型对重复样本产生显著局部偏好。为解决该问题，我们提出基于交叉熵的训练数据剪枝策略，用于去除过度相似的信号片段。同时，为弥补训练数据量减少带来的影响，我们通过生成含噪样本并应用基于KL散度的平滑诱导正则化来缓解过拟合。实验证明，本文提出的数据剪枝与正则化策略能带来稳定增益，且本框架在低资源场景下显著优于现有最优方法。