增量元学习IDBD算法在轴频电场信号检测中的应用  

Application of incremental meta-learning IDBD algorithm in signal detection of shaft-rate electric field

作　　者：卞强[1] 曾文仕 欧阳华[1] 童余德[1] BIAN Qiang;ZENG Wenshi;OUYANG Hua;TONG Yude(College of Electric Engineering,Naval University of Engineering,Hubei 430033,China)

机构地区：[1]海军工程大学电气工程学院,湖北武汉430033

出　　处：《国防科技大学学报》2022年第6期103-108,共6页Journal of National University of Defense Technology

基　　金：基础加强计划技术领域基金资助项目(2019-JCJQ-JJ-050)。

摘　　要：为了提高海洋环境电场背景中微弱舰船轴频电场的检测能力,针对传统的最小均方误差算法进行了改进,提出了一种基于增量元学习IDBD算法的自适应线谱增强器。利用所提算法对舰船缩比模型产生的实测轴频电场信号数据进行处理,结果表明该算法在低信噪比的情况下能够有效地将微弱轴频电场信号从宽带背景噪声中分离出来。所提算法相比于普通的自适应线谱算法,在改善信号的信噪比方面效果更加显著,且具有更快的收敛速度和更小的稳态误差,极大提高了舰船轴频电场的检测能力。In order to improve the detection ability of weak ship shaft-rate electric field in the background of marine environment electric field, the ALE(adaptive line enhancement) based on incremental meta-learning IDBD(incremental delta-bar-delta) algorithm was proposed to improve the traditional LMS(least mean square) algorithm. The proposed algorithm was used to process the measured shaft-rate electric field signal data generated by the ship scale model. The results show that the algorithm can effectively separate the weak shaft-rate electric field signal from the broadband background noise under the condition of low SNR(signal-to-noise ratio). Compared with the ordinary ALE algorithm, the proposed algorithm has a more significant effect in improving the SNR of the signal, and has a faster convergence speed and a smaller steady-state error, which greatly improves the ability to test shaft-rate electric field of the ship.

关 键 词：轴频电场 增量元学习 自适应 线谱增强 

分 类 号：TB559[理学—物理]