﻿ 基于大数据优化神经网络的船舶通信网络干扰信息识别
 舰船科学技术  2022, Vol. 44 Issue (14): 133-136    DOI: 10.3404/j.issn.1672-7649.2022.14.028 PDF

1. 中南大学，湖南 长沙 410007;
2. 湖南工业职业技术学院，湖南 长沙 410208;
3. 湖南安全技术职业学院，湖南 长沙 410151

Disturbance information identification of ship communication network based on big data optimization neural network
TAN Shao-sheng1,2, XIA Xu3
1. Central South University, Changsha 410007, China;
2. Hunan Industry Polytechnic, Changsha 410208, China;
3. Hunan Vocational Institute of Safety Technology, Changsha 410151, China
Abstract: Construct the characteristic recognition matrix of ship communication network interference information, select the instantaneous characteristic index of interference information, and extract characteristic information. The unipolar Sigmoid function neural network is used to construct the ship communication network interference information identification model, and the extracted instantaneous features are used as the input of the model. The weights between the hidden layer and the output layer are determined by the direct weight determination method, and the results of ship communication network interference information identification under the scheduling of big data resources are output. The results show that this method has a good recognition effect and can improve the security and transmission efficiency of communication information.
Key words: data resource scheduling     communication network     interference information identification     feature recognition matrix     neural network     Sigmoid function
0 引　言

1 船舶通信网络干扰信息识别方法 1.1 船舶通信网络大数据资源调度

 $RM = \left\langle {{E_Q}\left\langle {{E_D}} \right\rangle ,{S_M},{R_P}\left\langle {{R_D}} \right\rangle } \right\rangle 。$ (1)

1.2 大数据资源调度下船舶通信网络干扰信息特征采集 1.2.1 船舶通信网络干扰信息特征识别矩阵构建

 $C = \frac{{\displaystyle\sum\limits_i {\displaystyle\sum\limits_j {\left| {x\left( {i,j} \right)} \right|} } }}{{ZO}} 。$ (2)

1.2.2 干扰信息瞬时特征提取

 $K = \sqrt {R{\phi ^2} + B{\mu ^2}\left( {F\left( \partial \right)} \right)} 。$ (3)

 $\bar K = \frac{{N \times J\left( \partial \right)}}{{{K^2}}} 。$ (4)

 $\xi = \frac{1}{N} \times \bar K\left( {X\left( \varepsilon \right)} \right)。$ (5)

1.3 基于神经网络的干扰信息识别模型构建

$\left\{ {\left( {{X_i},{Y_i}} \right)} \right\}_{i = 1}^G$ 表示样本集内的 $G$ 个样本数据，利用式（6）表示输入样本 ${X_i}$ （即大数据资源调度下船舶通信网络内干扰信息的瞬时特征）、输出样本 ${Y_i}$ （即船舶通信网络内干扰信息识别结果）和神经网络的网络输出 ${\theta _i}$

 $\left\{ \begin{gathered} {X_i} = {\left[ {{x_{i1}},{x_{i2}}, \cdots {x_{im}}} \right]^{\rm{T}}} \in {R^m}，\\ {Y_i} = {\left[ {{y_{i1}},{y_{i2}}, \cdots {y_{in}}} \right]^{\rm{T}}} \in {R^n}，\\ {\theta _i} = {\left[ {{\theta _{i1}},{\theta _{i2}}, \cdots {\theta _{in}}} \right]^{\rm{T}}} \in {R^n}。\\ \end{gathered} \right.$ (6)

 $AE = \frac{{\left( {\displaystyle\sum\limits_{i = 1}^G {{{\left\| {{Y_i} - {\theta _i}} \right\|}^2}} } \right)}}{{Gn}} 。$ (7)

 $F = \left[ \begin{array}{*{20}{c}} {f_{11}}&{f_{12}}&{ \cdots}&{f_{1k}} \\ {f_{21}}&{f_{22}}&{ \cdots }&{f_{2k}}\\ { \vdots }&{ \vdots }&{}& \vdots \\ {f_{N1}}&{f_{N2}}&{ \cdots }&{f_{Nk}} \\ \end{array} \right]。$ (8)

 $Y = \left[ \begin{array}{*{20}{c}} {y_{11}}&{y_{12}}&{ \cdots }&{y_{1n}} \\ {y_{21}}&{y_{22}}&{ \cdots }&{y_{2n}} \\ { \vdots }&{ \vdots}&{}& \vdots \\ {y_{N1}}&{y_{N2}}&{ \cdots }&{y_{Nn}} \\ \end{array} \right] = {R^{n \times N}}。$ (9)

 $\begin{split} W =& \left[ {{W_1},{W_2}, \cdots ,{W_n}} \right] = \\ & \left[ \begin{array}{*{20}{c}} {w_{11}}&{w_{12}}&{ \cdots}&{w_{1n}} \\ {w_{21}}&{w_{22}}&{ \cdots}&{w_{2n}} \\ { \vdots}&{ \vdots }&{}& \vdots \\ {w_{k1}}&{w_{k2}}&{ \cdots }&{w_{kn}} \\ \end{array} \right] = {R^{n \times k}} \end{split}。$ (10)

 $W = {F^ + }Y 。$ (11)

2 研究结果 2.1 干扰信息识别结果

 图 1 低频干扰信息识别结果 Fig. 1 Identification results of low frequency interference information

 图 3 高频干扰信息识别结果 Fig. 3 Identification results of high frequency interference information

 图 2 中频干扰信息识别结果 Fig. 2 Identification results of if interference information
2.2 神经元数量对于干扰信息识别结果的影响

2.3 通信数据的同步率

2.4 通信信息传输效率分析

 图 4 信息传输效率的对比结果 Fig. 4 Comparison results of information transmission efficiency

3 结　语

 [1] 彭忠益, 卢珊. 大数据驱动的国家矿产资源安全管理信息平台模型研究[J]. 情报杂志, 2021, 40(12): 163-168+202. [2] 陈江兴, 梁良, 付俊峰, 等. 基于大数据的智能电网数据调度与快速分发方法研究[J]. 电测与仪表, 2020, 57(6): 88-93. [3] 谢添, 高士顺, 赵海涛, 等. 基于强化学习的定向无线通信网络抗干扰资源调度算法[J]. 电波科学学报, 2020, 35(4): 531-541. [4] 罗彬珅, 刘利民, 董健, 等. 基于SAE-GA-SVM模型的雷达新型干扰识别[J]. 计算机工程, 2020, 46(6): 281-287. [5] 陈涛, 刘姗姗, 梁修荣. 一种新的无线传感网络干扰攻击检测技术[J]. 计算机应用研究, 2020, 37(12): 3757-3759. [6] 李旭, 姚春龙, 范丰龙, 等. 结合注意力机制的循环神经网络复述识别模型[J]. 控制与决策, 2021, 36(1): 152-158.