﻿ 基于WLAN的大型船舶舱体内部指纹定位算法
 舰船科学技术  2018, Vol. 40 Issue (8): 114-118 PDF

WLAN-based fingerprint location algorithm for large vessel
XU Zhi-xun, GAO Shang
School of Computer Science and Engineering, Jiangsu University of Science and Technology, Zhenjian 212003, China
Abstract: Large internal structure of the ship is complex, Traditional navigation and positioning technology can not penetrate the cabin to achieve staff positioning. This paper designs the RSS fingerprint positioning method based on WLAN. In the offline phase, collect and train RSS location fingerprints. In the online positioning phase, the data collected in real time is compared with the offline fingerprint library to estimate the positioning result. Aiming at the characteristics of complex and closed space of the ship's cabin, the Bayesian clustering algorithm and the algorithm of AP selection based on information gain are proposed in the online and offline stages of positioning, and the simulation experiment is carried out with the traditional algorithm Compared. The results show that the proposed algorithm can meet the accuracy requirements of positioning in the cabin, and further improve the positioning accuracy compared with the original algorithm.
Key words: cabin positioning     location fingerprint     positioning algorithm     AP selection algorithm
0 引　言

1 WLAN指纹定位模型

WLAN是一种基于IEEE802.11标准的无线局域网业务。在WLAN的网络覆盖区域内，若干接入点（AP）分散于空间各处，其信号的收发符合无线电磁波的传播规律，可以利用数学模型将传播信号转化为确定性的长度、角度等参数，或者概率性的指纹特征参数。

 图 1 WLAN指纹定位流程 Fig. 1 WLAN fingerprint positioning process

 $RS\!S = {P_t} - K - 10\alpha {\log _{10}}d{\text{，}}$ (1)

1.1 离线训练阶段

1.2 在线估算阶段

 图 4 在线匹配估算 Fig. 4 Online match estimate

2 贝叶斯分簇定位算法

 $P\left( {L_{i}\left| {RSS'} \right.} \right) = \frac{{P\left( {RSS'\left| {L_{i}} \right.} \right) \cdot P\left( {L_{i}} \right)}}{{P\left( {RSS'} \right)}}{\text{，}}$ (2)

 ${P_{sum}}{{ = }}\sum\limits_{i = 1}^{{N_c}} {{P_i} = \sum\limits_{i = 1}^{{N_c}} {\sum\limits_{j = 1}^{{N_i}} {{P_{ij}}} } } {\text{，}}$ (3)

 ${L_x} = \frac{{\sum\limits_{t = 1}^{{N_c}} {P\left( {RSS'\left| {{L_t}{l_t}} \right.} \right)} }}{{\sum\limits_{t = 1}^{{N_c}} {P\left( {RSS'\left| {{L_t}} \right.} \right)} }}{\text{。}}$ (4)
3 基于信息增益的AP选择改进算法 3.1 算法流程

 图 5 AP选择流程图 Fig. 5 AP selection flow chart

3.2 信息增益计算

 $\begin{array}{l}InfoGain\left( {A{P_i}} \right) = H\left( L \right) - H\left( {L|A{P_i}} \right){\text{，}}\\{\kern 1pt} \;\;\;\;\;\;\left( {1 \leqslant i \leqslant m} \right){\text{。}}\end{array}$ (5)

 $H\left( L \right) = - \sum\nolimits_{j = 1}^c {Pr \left( {{L_j}} \right)\operatorname{log}Pr } \left( {{L_j}} \right){\text{，}}$ (6)
 $H\left( {L|A{p_i}} \right) \!=\!\! -\!\! \sum\nolimits_v {\sum\nolimits_{j = 1}^c {Pr \left( {{L_j},A{P_i}v} \right)\operatorname{log } Pr} \left( {{L_j}|A{P_i} \!=\! v} \right)} {\text{。}}$ (7)

3.3 综合可辨识性计算

 $D(A{P_i}) = InfoGain(A{P_i}) + r \times F(A{P_i}){\text{，}}$ (8)
 $F(A{P_i}) = \sum\nolimits_{m = 1}^n {\frac{{\left| {Rs{s_i} - Rs{s_m}} \right|}}{{\left| {{L_i} - {L_m}} \right|}}} {\text{。}}$ (9)

4 仿真与实验结果分析

 图 6 定位算法比较 Fig. 6 Positioning algorithm comparison

 图 7 AP选择算法定位误差 Fig. 7 AP selection algorithm positioning error

 图 8 误差累积概率 Fig. 8 Error accumulation probability

3种AP选择算法下的累积误差分布图如图8所示。可以看出，本文的AP选择算法使整体定位精度明显提升，同InfoGain算法和Max-Mean算法相比，定位误差在1.5 m以内的概率分别增加了12%和24%；定位误差在2 m以内的概率分别增加了7%和13%。

5 结　语

 [1] 苑隆寅. 基于WLAN的船舶通信网络应用与仿真[J]. 舰船科学技术, 2015, (01): 191–195. YUAN Long-yin. The application and simulation of shipboard communication network based on WLAN[J]. Ship Science and Technology, 2015, (01): 191–195. http://cdmd.cnki.com.cn/Article/CDMD-10269-2007082219.htm [2] 王玲, 张彬祥. 船舶通信导航技术及发展趋势[J]. 舰船电子工程, 2016, (03): 17–21. WANG Ling, ZHANG Bin-xiang. Development trend of marine communication and navigation technology[J]. Ship Electronic Engineering, 2016, (03): 17–21. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=shzc200903001 [3] 屈召贵, 刘强. 基于北斗卫星的航行数据导航方法研究[J]. 舰船科学技术, 2016, (08): 145–147. QU Zhao-gui, LIU Qiang. Research on navigation data navigation method based on Beidou satellite[J]. Ship Science and Technology, 2016, (08): 145–147. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jdjs201206055 [4] 马远美. 船舶导航雷达运动目标跟踪性能改进方法研究[D]. 大连: 大连海事大学, 2013. [5] 陈迅. 无线传感器网络通信协议及定位算法研究[D]. 上海: 复旦大学, 2007. [6] 钱志鸿, 孙大洋, LEUNG Victor. 无线网络定位综述[J]. 计算机学报, 2016, (06): 1237–1256. QIAN Zhi-hong, SUN Da-yang, LEUNG Victor. A survey on localization model in wireless networks[J]. Chinese Journal of Computers, 2016, (06): 1237–1256. [7] 辛艳, 梁建坤, 修长虹. 无线局域网IEEE802.11ac协议的研究及应用[J]. 网络安全技术与应用, 2016, (09): 80–81. XIN Yan, LIANG Jian-kun, XIU Chang-hong. Research and application of wireless LAN IEEE802.11ac protocol [J]. Network Security Technology & Application, 2016, (09): 80–81. [8] 丁根明. 基于人工智能的室内指纹定位技术研究[D]. 北京: 北京交通大学, 2015. [9] 梁峰. 大型船舶内部定位导航方法[J]. 中国科技信息, 2016, (21): 60–61. LIANG Feng. Navigation method of large ship's internal location[J]. China Science and Technology Information, 2016, (21): 60–61. [10] 李言胜, 孙琳, 王亚坤. 基于ZigBee的舱内人员巡检系统的应用研究[J]. 电脑知识与技术, 2016, (26): 238–239+245. LI Yan-sheng, SUN Lin, WANG Ya-kun. Application of Operator Patrol System Based on ZigBee in the Cabin[J]. Computer Knowledge and Technology, 2016, (26): 238–239+245. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dnzsyjs-itrzyksb201626100 [11] 叶宝玉, 王钦若, 等. 基于超声波的模型船舶室内定位系统研究[J]. 计算机工程, 2012, (19): 258–260+265. YE Bao-yu, WANG Qin-ruo, et al. Research on Indoor Location Systems for Ship Model Based on Ultrasonic[J]. Computer Engineering, 2012, (19): 258–260+265. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjgc201219066 [12] 谢泰, 刘晓荣, 等. 舰船人员舱内定位及生命体征监测系统研究[J]. 医疗卫生装备, 2016, (03): 11–13. XIE Tai, LIU Xiao-rong, et al. Research on inboard positioning and vital signs monitoring system for ship crews[J]. Chinese Medical Equipment Journal, 2016, (03): 11–13. http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_ylwszb201603004 [13] 胡明成. WLAN中集中式的AP选择机制和节能调度方法研究[D]. 西安电子科技大学, 2015. [14] 陈淼. 基于信号强度的WLAN室内定位跟踪系统研究[D]. 武汉: 武汉大学, 2012. [15] 徐旻捷. 可扩展的贝叶斯学习方法建模与推理[D]. 北京: 清华大学, 2015. [16] YOUSSEF M, AGRAWALA A. Handling samples correlation in the horus system[C]//IEEE InfoCom 2003. 2004(2): 1023–1031. [17] CHEN K Y, YANG Q, YIN J, et al. Power-Efficient Access-Point Selection for Indoor Location Estimation[J]. IEEE Transactions on Knowledge and Data Engineering. 2006, 18(7): 877–888.