加权有向关联网络构建与表征的水中目标远距离检测

Construction and Characterization of Weighted Directed Association Network for Remote Detection of Underwater Targets

水中目标的远距离检测是海洋防御体系的关键技术之一,对国防及民用领域均具有十分重要的作用。然而,目前尚缺乏行之有效的水中目标远距离检测方法,特别是目标先验信息未知的情况下变的愈加困难。为解决这一问题,提出一种新的方法—加权有向关联网络。通过矢量声信号到加权有向关联网络的映射,将信号检测问题转化为网络拓扑的表征,并通过对网络拓扑的特性分析及特征提取,实现无目标先验信息下的水中目标远距离检测。并通过仿真与实测数据对所提出的方法进行验证。研究结果表明:与现有的窄带互谱检测、冒泡熵等方法相比,所提方法能够检测到更低信噪比的水中目标,实现了无需目标先验信息的水中目标远距离检测;该方法的应用具有一定的实际意义和应用前景,可以为海洋防御和民用领域的水下目标检测提供有效的技术支撑。

Remote detection of underwater targets is a critical technology in ocean defense systems and has significant applications in both national defense and civilian domains.However,there is currently a lack of effective methods for long-range detection of underwater targets,especially without prior knowledge of the target.To address this issue,a new detection method based on weighted directed association network,is proposed,which represents the target signal detection problem as network topology by mapping the vector acoustic signal into the weighted directed association network.The long-range detection of underwater targets without prior knowledge is achieved through the analysis of network topology characteristics and the feature extraction.The proposed method is validated through the simulated and measured data.Compared with existing narrowband mutual spectrum detection and bubble entropy methods,the proposed method can detect the underwater targets with lower signal-to-noise ratios and achieve the long-range detection without prior knowledge.The proposed method has important practical significance and application prospects and can provide the effective technical support for underwater target detection in ocean defense and civilian fields.