State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, 92 Weijin Road, Nankai District, Tianjin, China.
China Shipbuilding System Engineering Research Institute, Beijing 100036, China.
J Acoust Soc Am. 2022 Jul;152(1):491. doi: 10.1121/10.0012693.
Bionic camouflage covert underwater acoustic communication has recently attracted great attention. However, we have not found relevant methods or literature to recognize these bionic camouflage communication signals (BCCSs) in the area of anti-reconnaissance. Focused on recognizing the BCCSs, this article proposes a recognition method based on the statistics of inter-click intervals to recognize the camouflaged click communication train (CCCT), which is modulated by time delay difference (TDD). We first analyze the characteristics of TDD distributions of CCCT and real click train (RCT). According to the coding principle, the TDDs of CCCTs present a ladder-like distribution with a fixed time step, and the TDDs are equal to the integral multiple of the fixed time step. On the contrary, the TDDs of RCTs are approximately random distribution within a certain time range. Therefore, based on the different TDD distributions, this article classifies CCCTs and RCTs by utilizing the statistical property of TDD distributions. To measure the TDDs of diverse cetacean clicks accurately, a new click location scheme based on the dynamic window energy ratio is proposed. Next, based on the statistics of TDD distribution, the influences of the TDDs that are caused by multipath interferences are eliminated by iteration. Simulations demonstrate the accuracy of the recognition method under different conditions.
仿生伪装隐蔽水下声通信最近引起了广泛关注。然而，在反侦察领域，我们尚未找到相关方法或文献来识别这些仿生伪装通信信号（BCCSs）。本文针对识别 BCCSs 的问题，提出了一种基于点击间隔统计的识别方法，用于识别由延时差（TDD）调制的伪装点击通信序列（CCCT）。我们首先分析了 CCCT 和真实点击序列（RCT）的 TDD 分布特征。根据编码原理，CCCT 的 TDD 呈现出固定时间步长的阶梯状分布，并且 TDD 等于固定时间步长的整数倍。相反，RCT 的 TDD 在一定时间范围内近似随机分布。因此，本文基于 TDD 分布的不同，利用 TDD 分布的统计特性对 CCCT 和 RCT 进行分类。为了准确测量各种鲸类点击的 TDD，我们提出了一种基于动态窗口能量比的新点击定位方案。然后，基于 TDD 分布的统计特性，通过迭代消除由多径干扰引起的 TDD 影响。仿真结果表明，在不同条件下该识别方法的准确性。
