﻿ 船岸通信技术下舰船导航信号非线性滤波
 舰船科学技术  2022, Vol. 44 Issue (20): 139-142    DOI: 10.3404/j.issn.1672-7649.2022.20.028 PDF

Nonlinear filtering of ship navigation signal based on ship shore communication technology
WANG Yong-jie
Department of Information Engineering, Tianjin Maritime College, Tianjin 300350, China
Abstract: Ship navigation signal is easily disturbed by noise, so a nonlinear filtering method of ship navigation signal based on ship shore communication technology is proposed. Calculate the state trajectory parameters of the normal operation of the ship, build a first-order Markov normalization equation, and make different types of noise signals into different nonlinear sequences, and complete the navigation signal denoising through discrete processing. The measurement model is built to reduce the navigation signal reception deviation. The ship shore communication technology is used to calculate the transmission delay of noise signals and ordinary signals in the sequence. The covariance matrix is introduced to solve the mean value of the signal delay through the matrix. The nonlinear filtering is realized by constraining the delay of the navigation channel. The experimental data show that the nonlinear filtering effect of the proposed method is good, and it can achieve efficient and stable filtering for noise signals of different intensity. Therefore, the proposed method has high overall practical value and strong applicability.
Key words: ship shore communication technology     nonlinear filtering     noise signal     nonlinear sequence     transmission delay
0 引　言

1 舰船噪声信号去噪

 $\begin{gathered} \phi = {v_{CN}} + v\cos H + {w_1}，\\ \lambda = {v_{CE}} + v\sin H + {w_2}，\\ v = {w_5}，\\ H = \Omega + {w_6}，\\ \Omega = {w_7}。\\ \end{gathered}$ (1)

 $\begin{gathered} {\xi_{CE}} = - \beta v_{CE} + {w_3}，\\ {\xi_{CN}} = - \beta v_{CN} + {w_4} 。\\ \end{gathered}$ (2)

 $\begin{split} \phi \left( k \right) = &\phi \left( {k - 1} \right) + \left( {1 - {e^{ - \beta }}} \right)/{\beta _c}{v_{CN}}\left( {k - 1} \right) - v\left( {k - 1} \right) \times \\ &\cos \left( {H\left( {k - 1} \right) + T/2\Omega \left( {k - 1} \right)} \right)T + {w_1}\left( {k - 1} \right) ，\end{split}$ (3)
 $\begin{split} \lambda \left( k \right) =& \lambda \left( {k - 1} \right) + \left( {1 - {e^{ - \beta }}} \right)/{\beta _c}{v_{CE}}\left( {k - 1} \right) - v\left( {k - 1} \right) \times \\ & \sin \left( {H\left( {k - 1} \right) + T/2\Omega \left( {k - 1} \right)} \right)T + {w_2}\left( {k - 1} \right) 。\end{split}$ (4)

 $\begin{gathered} {v_{CE}}\left( k \right) = {e^{ - \beta }}{v_{CE}}\left( {k - 1} \right) + {w_3}\left( {k - 1} \right) ，\\ {v_{CN}}\left( k \right) = {e^{ - \beta }}{v_{CN}}\left( {k - 1} \right) + {w_4}\left( {k - 1} \right)，\\ v\left( k \right) = v\left( {k - 1} \right) + {w_5}\left( {k - 1} \right)，\\ H\left( k \right) = H\left( {k - 1} \right) + \Omega \left( {k - 1} \right)T + {w_6}\left( {k - 1} \right)，\\ \Omega \left( k \right) = \Omega \left( {k - 1} \right) + {w_7}\left( {k - 1} \right)。\\ \end{gathered}$ (5)

 $X\left( k \right) = f\left( {X\left( {k - 1} \right),k - 1} \right) + \Gamma W\left( {k - 1} \right) 。$ (6)

2 舰船导航系统信号量测模型

 $z\left( t \right) = h\left( t \right)x\left( t \right) + v\left( t \right)，$ (7)

 $\begin{gathered} z = \left[ \begin{gathered} \left( {{L_I} - {L_M}} \right){R_M} \\ \left( {{\gamma _I} - {\gamma _M}} \right){R_N}\cos L \\ {h_I} - {h_M} \\ \end{gathered} \right] = \left[ \begin{gathered} {R_M}\varsigma L \\ {R_N}\cos L\varsigma \gamma \\ \varsigma h \\ \end{gathered} \right] = \left[ \begin{gathered} {v_1} \\ {v_2} \\ {v_3} \\ \end{gathered} \right]，\\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} = {\kern 1pt} {\kern 1pt} \left[ {\begin{array}{*{20}{c}} {}&{{R_M}}&{\begin{array}{*{20}{c}} 0&{{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} 0}&{} \end{array}} \\ {{0_3} \times 7}&0&{\begin{array}{*{20}{c}} {{R_N}\cos L}&0&{{0_3} \times 9} \end{array}} \\ {}&0&{\begin{array}{*{20}{c}} 0&{{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} 1}&{} \end{array}} \end{array}} \right]。\\ \end{gathered}$ (8)

3 导航信号非线性滤波

 $\begin{gathered} {x_{l + 1}} = {f_l}\left( {{x_l},{u_l},{w_l}} \right)，\\ {y_l} = {g_l}\left( {{x_l},{\upsilon _l}} \right) 。\\ \end{gathered}$ (9)

 $\begin{gathered} E\left( {{w_l}} \right) = {{w''}_l},E\left[ {\left( {{w_l} - {{w''}_l}} \right){{\left( {{w_l} - {{w''}_l}} \right)}^{\rm{T}}}} \right] = {Q_l}，\\ E\left( {{\upsilon _l}} \right) = {{\upsilon ''}_l},E\left[ {\left( {{\upsilon _l} - {\upsilon _l}} \right){{\left( {{\upsilon _l} - {\upsilon _l}} \right)}^{\rm{T}}}} \right] = {F_l} 。\\ \end{gathered}$ (10)

 $\begin{gathered} {Q_l} = {G_w}G_w^{\rm{T}},{F_l} = {G_\upsilon }G_\upsilon ^{\rm{T}} ，\\ {{\bar P}_l} = {{\bar G}_w}\bar G_w^{\rm{T}},{{\bar P}_l} = {{\bar G}_\upsilon }\bar G_\upsilon ^{\rm{T }}。\\ \end{gathered}$ (11)

 \left\{ \begin{aligned} &{{\bar y}_{l + 1}} = g\left( {{{\bar x}_{l + 1}},{{\bar \upsilon }_{l + 1}}} \right)，\\ &K{{\bar x}_{l + 1}} = {Q_x}Q_{xy}^{\rm{T}}\left[ {{Q_{yx}}Q_{yx}^{\rm{T}} + {F_{yx}}F_{yx}^{\rm{T}}} \right]，\\ &{{\hat x}_{l + 1}} = {{\bar x}_{l + 1}} + K{{\bar x}_{l + 1}}\left[ {{y_{l + 1}} - {{\bar y}_{l + 1}}} \right]，\\ &{{\hat P}_{l + 1}} = \left[ {{Q_x} - K{{\bar x}_{l + 1}}{Q_{yx}}{Q_{y\upsilon }}} \right] \times {\left[ {{Q_x} - K{{\bar x}_{l + 1}}{Q_{yx}}{K_{l + 1}}{Q_{y\upsilon }}} \right]^{\rm{T}}} 。\\ \end{aligned} \right. (12)

4 滤波性能测试 4.1 实验背景

4.2 非线性滤波前后导航信号噪声变化

 图 1 –10 dB噪声强度下降噪前后信号变化 Fig. 1 Signal change before and after –10 dB noise intensity reduction

 图 2 10 dB噪声强度下降噪前后信号变化 Fig. 2 Signal change before and after 10 dB noise intensity reduction

5 结　语

 [1] 汪昭河, 黄蕾, 王艳杰, 等. UFMC系统中基于Volterra滤波器的非线性失真补偿方案[J]. 微电子学与计算机, 2020, 37(1): 1-6+13. DOI:10.19304/j.cnki.issn1000-7180.2020.01.001 [2] 陈振炜, 孟义朝, 詹遥牧. 延时可变光电振荡器产生混沌信号的路径及其特性研究[J]. 激光与光电子学进展, 2020, 57(19): 211-220. [3] 王领, 申晓红, 张之琛, 等. 一种适用于水声移动通信同步检测的组合滤波器[J]. 西北工业大学学报, 2020, 38(5): 919-927. DOI:10.3969/j.issn.1000-2758.2020.05.001 [4] 孙青丰, 李大为, 王韬, 等. 基于自相关的纳焦级弱信号10~(11)高动态范围测量[J]. 中国激光, 2020, 47(6): 175-182. [5] 张嘉纹, 党小宇, 杨凌辉, 等. 海面短波地波通信中基于DNN神经网络的单样本极化滤波器预测研究[J]. 电子学报, 2020, 48(11): 2250-2257. DOI:10.3969/j.issn.0372-2112.2020.11.022 [6] 陈霖, 周廷, 刘占超. GNSS失锁下基于混合预测模型的POS误差估计方法[J]. 中国惯性技术学报, 2022, 30(1): 74-80. DOI:10.13695/j.cnki.12-1222/o3.2022.01.011 [7] 李远禄, 赵伟静, 蒋民. 基于时间分数阶非线性扩散模型的平滑方法[J]. 计算机应用研究, 2020, 37(3): 704-707. [8] 刘长德, 顾宇翔, 张进丰. 基于小波滤波和LSTM神经网络的船舶运动极短期预报研究[J]. 船舶力学, 2021, 25(3): 299-310. DOI:10.3969/j.issn.1007-7294.2021.03.005 [9] SPOTTS I, BRODIE C H, GADSDEN S A, et al. Comparison of nonlinear filtering techniques for photonic systems with blackbody radiation[J]. Applied Optics, 2020, 59(30): 9303-9312. DOI:10.1364/AO.403484 [10] 马月红, 张伟涛, 惠蕙, 等. 基于单比特接收机的高检测率测频方法设计与实现[J]. 北京理工大学学报, 2021, 41(7): 774-780. DOI:10.15918/j.tbit1001-0645.2020.225 [11] 吴健, 孙永波, 赵前进. 基于神经网络的周期扰动非线性系统自适应渐近跟踪控制[J]. 控制与决策, 2022, 37(4): 922-932. DOI:10.13195/j.kzyjc.2020.1252 [12] 刘素贞, 魏建, 张闯, 等. 基于FPGA的超声信号自适应滤波与特征提取[J]. 电工技术学报, 2020, 35(13): 2870-2878. DOI:10.19595/j.cnki.1000-6753.tces.190572 [13] 赵修平, 齐嘉兴, 崔伟成, 等. MOMEDA结合数学形态滤波的齿轮故障特征提取[J]. 机械科学与技术, 2020, 39(2): 247-252. DOI:10.13433/j.cnki.1003-8728.20190122 [14] 魏志强, 程勇策, 辛林杰, 等. 非线性跟踪微分器在光电吊舱跟踪系统中的应用研究[J]. 电视技术, 2020, 44(4): 39-43. DOI:10.16280/j.videoe.2020.04.009