﻿ 决策树算法在无人船异构通信网络切换中的应用
 舰船科学技术  2022, Vol. 44 Issue (24): 67-70    DOI: 10.3404/j.issn.1672-7649.2022.24.014 PDF

Application of decision tree algorithm in heterogeneous communication network switching of unmanned ship
HUA Zhen-xing
Department of Computer and Information Science, Boda College of Jilin Normal University, Siping 136000, China
Abstract: Aiming at the problem of poor communication quality of heterogeneous communication network due to the complicated navigation environment of unmanned ship, the application of decision tree algorithm in the switching of heterogeneous communication network of unmanned ship is studied. The signal strength, maximum transmission rate and bit error rate were selected as the network attributes of the unmanned ship heterogeneous communication network switching, and each network attribute was set as the decision node of the C4.5 decision tree algorithm. The C4.5 decision tree algorithm selected the information gain rate as the attribute classification standard to construct the candidate network set. The communication attribute values in the candidate network set were normalized to obtain the multi-attribute decision values of each communication network. The network with the most attribute decision values was selected as the target network to complete network switching. The experimental results show that this method can effectively switch the heterogeneous communication network of unmanned ship, and the network throughput is higher than 50 Mbps.
Key words: decision tree algorithm     unmanned ship     heterogeneous communication     network switching     signal strength     bit error rate
0 引　言

1 无人船异构通信网络切换 1.1 无人船异构通信网络结构

 图 1 无人船异构通信网络 Fig. 1 Heterogeneous communication network of unmanned ship

1.2 无人船异构通信网络属性分析

1）信号强度

 $RSS = G - K\lg \left( d \right) + v\left( x \right) 。$ (1)

 ${P_1} = P\left( {RS{S_B} > \varphi } \right)。$ (2)

2）误码率

 $SNR = RSS/I，$ (3)

 $BER = Q\sqrt {SNR}，$ (4)

 $Q\left( x \right) = \dfrac{1}{{\sqrt {2 \text{π} } }}\int\nolimits_x^\infty {\exp \left( { - {t^2}/2} \right){\rm{d}}t}$ (5)

 ${P_2} = P\left( {BER' < \delta } \right)。$ (6)

3）最大传输速率

 $L = W\left( {1 + SNR} \right)，$ (7)

 ${P_3} = P\left( {L' > L} \right)。$ (8)

1.3 基于C4.5决策树算法的候选网络集生成

1）计算样本熵值

 $e\left( B \right) = - \sum\limits_{j = 1}^{\left| C \right|} {Pr \left( {{c_j}} \right){{\log }_2}} Pr \left( {{c_j}} \right) 。$ (9)

2）计算信息增益值

 $e\left( {A,B} \right) = \sum\limits_{j = 1}^v {\frac{{\left| {{B_j}} \right|}}{B}} \times e\left( {{B_j}} \right)，$ (10)

 $Gain\left( {B,A} \right) = e\left( B \right) - e\left( {A,B} \right)。$ (11)

3）计算信息增益比

 $GainRadio\left( {B,A} \right) = \frac{{Gain\left( {B,A} \right)}}{{ - \displaystyle\sum\limits_{j = 1}^s {\left( {\frac{{{B_j}}}{B} \times {{\log }_2}\frac{{\left| {{B_j}} \right|}}{{\left| B \right|}}} \right)} }}。$ (12)

1.4 基于决策树的无人船异构通信网络切换

 ${F_i}\left( x \right) = \frac{{RS{S_i}}}{{{\xi _i}}}\left( {1 - {P_1}} \right)\left( {1 - {P_2}} \right)\left( {1 - {P_3}} \right) ，$ (13)

 ${F_k}\left( x \right) = \max \left\{ {{F_1}\left( x \right),{F_2}\left( x \right), \cdots ,{F_n}\left( x \right)} \right\} ，$ (14)

 图 2 无人船异构通信网络切换流程图 Fig. 2 Flowchart of heterogeneous communication network switching of unmanned ship

2 仿真测试与分析

 图 3 仿真实验场景 Fig. 3 Simulation experiment scene

 图 4 不同通信网络的信号强度 Fig. 4 Signal strength of different communication networks

 图 5 无人船异构网络吞吐量变化 Fig. 5 Throughput changes of heterogeneous network of unmanned ships
3 结　语

 [1] 郭艳卿, 王鑫磊, 付海燕, 等. 面向隐私安全的联邦决策树算法[J]. 计算机学报, 2021, 44(10): 2090-2103. GUO Yanqing, WANG Xinlei, FU Haiyan, et al. Federated Decision Tree Algorithm for Privacy Security[J]. Chinese Journal of Computers, 2021, 44(10): 2090-2103. [2] 安葳鹏, 程小博, 刘雨. Fleiss' Kappa系数在贝叶斯决策树算法中的应用[J]. 计算机工程与应用, 2020, 56(7): 137-140. AN Weipeng, CHENG Xiaobo, LIU Yu. Application of Fleiss 'Kappa coefficient to Bayesian decision tree algorithm[J]. Computer Engineering and Applications, 2020, 56(7): 137-140. [3] 陈立家, 周为, 许毅, 等. 一种基于SDN的多约束无人船网络传输路由算法[J]. 中国舰船研究, 2022, 17(4): 107-113. CHEN Lijia, ZHOU Wei, XU Yi, et al. Multi-constrained unmanned surface vessel network transmission routing algorithm based on SDN[J]. Chinese Journal of Ship Research, 2022, 17(4): 107-113. DOI:10.19693/j.issn.1673-3185.02454 [4] 王磊, 刘雨, 刘志中, 等. 基于属性离散和特征度量的决策树构建算法[J]. 河南理工大学学报(自然科学版), 2021, 40(3): 127-133. WANG Lei, LIU Yu, LIU Zhizhong, et al. Decision tree construction algorithm based on attribute dispersion and feature measurement[J]. Journal of Henan Polytechnic University(Natural Science), 2021, 40(3): 127-133. DOI:10.16186/j.cnki.1673-9787.2020040086 [5] 刘炜, 李东坤, 徐畅, 等. 应急通信网络中基于粒子群优化的信道分配算法[J]. 计算机科学, 2021, 48(5): 277-282. LIU Wei, LI Dongkun, XU Chang, et al. Channel Assignment Algorithm Based on Particle Swarm Optimization in Emergency Communication Networks[J]. Computer Science, 2021, 48(5): 277-282.