基于高斯似然的精准水声信道估计
Accurate underwater acoustic channel estimation based on Gaussian likelihood
杨光 1乔培玥 2梁俊燕 2秦正昌 2巩小东 3倪秀辉3
作者信息
- 1. 青岛理工大学信息与控制工程学院青岛市水声通信及探测装备技术创新中心,山东青岛 266525;南洋理工大学电气与电子工程学院,新加坡 639798
- 2. 青岛理工大学信息与控制工程学院青岛市水声通信及探测装备技术创新中心,山东青岛 266525
- 3. 山东省科学院海洋仪器仪表研究所,山东青岛 266318;乌克兰国立技术大学(基辅工学院),乌克兰基辅03056
- 折叠
摘要
针对时变水声信道的多途干扰问题,提出基于高斯似然(Gaussian likelihood,GL)的精准水声信道估计算法.GL算法将相邻信道短块的高斯概率密度函数相乘,乘积仍然服从高斯分布,且方差变小,从而进一步提高信道估计的准确性;采用叠加训练(superimposed training,ST)方案,将训练序列和符号序列线性叠加,使训练序列持续传输,实现对信道的实时跟踪.将ST方案、GL算法和Turbo均衡以迭代的方式相结合,估计出的符号序列作为虚拟训练序列,进一步提高时变水声信道的估计和跟踪性能.通过多次迭代计算,实现时变水声信道的精准估计和实时跟踪.最后,通过计算机仿真以及胶州湾收发节点水平距离500 m和5.5 km的海上运动实装试验,验证了所提算法的有效性.
Abstract
Aiming at the multi-path interference problem of the time-varying underwater acoustic channel,an accurate underwater acoustic channel estimation algorithm based on Gaussian likelihood(GL)is proposed.The Gauss probability density functions of the multiple segments are multiplied,the product result still follows the Gauss distribution,and the variance becomes smaller,leading to the improvement of channel estimation accuracy.The superimposed training(ST)scheme is used,where the training sequence and the symbol sequence are linearly superimposed,so that the training sequence can be continuously transmitted,thereby the real-time tracking of the time-varying channel is realized.The ST scheme,GL algorithm,and Turbo equalization are jointly performed in an iterative manner,where the estimated symbol sequence is used as a virtual training sequence to further improve the estimation and tracking performance of the channel.Accurate estimation and real-time tracking of the time-varying underwater acoustic channel are realized through multiple iteration calculation.Finally,the effectiveness of the proposed algorithm is verified by simulation and experimental results(the horizontal distance of the moving transceivers is 500 m and 5.5 km)in Jiaozhou Bay.
关键词
时变水声信道/高斯似然/叠加训练方案/虚拟训练序列Key words
time-varying underwater acoustic channel/Gaussian likelihood(GL)/superimposed training(ST)scheme/virtual training sequence引用本文复制引用
基金项目
国家自然科学基金(61771271)
山东省自然科学基金面上项目(ZR2020MF010)
山东省自然科学基金面上项目(ZR2020MF001)
青岛市源头创新计划-青年专项(19-6-2-4-cg)
青岛市关键技术攻关及产业化示范类项目(22-3-3-hygg-8-hy)
出版年
2024
NETL
NSTL
万方数据