[1]刘 伟,程春玲,杜金楷,等.一种基于改进深度卷积模型的室内定位方法[J].计算机技术与发展,2022,32(08):155-160.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 025]
　LIU Wei,CHENG Chun-ling,DU Jin-kai,et al.An Indoor Positioning Method Based on Improved Deep Convolution Mode[J].,2022,32(08):155-160.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 025]
点击复制

一种基于改进深度卷积模型的室内定位方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年08期
页码:
155-160
栏目:
应用前沿与综合
出版日期:
2022-08-10

文章信息/Info

Title:
An Indoor Positioning Method Based on Improved Deep Convolution Mode
文章编号:
1673-629X(2022)08-0155-06
作者:
刘 伟; 程春玲; 杜金楷; 章丁祥
南京邮电大学 计算机学院,江苏 南京 210003
Author(s):
LIU Wei; CHENG Chun-ling; DU Jin-kai; ZHANG Ding-xiang
School of Computer,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
关键词:
室内定位; 卷积神经网络; 开放集识别; 信道状态信息; 数据多维图像
Keywords:
indoor positioning; convolutional neural network; open set recognition; channel state information; data multi-dimensional image
分类号:
TP183
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 08. 025
摘要:
在室内定位中利用深度卷积神经网络(Deep Convolutional Neural Network,DCNN)对信道状态信息(Channel StateInformation,CSI)构建图像并实现位置分类是一种新的方法。 但基于 DCNN 的室内定位算法受实际环境、数据采集方式、模型结构等的影响,存在拟合能力弱、定位精度较差、未知位置识别能力弱的问题,在实际定位中很难满足定位精确度和稳定性的需求。 对此提出一个能够识别未知位置的改进模型:I-DCNN( Improved-Deep Convolutional Neural Network)。针对拟合能力弱的问题,通过增加 DCNN 的复杂度和调整模型结构的方法来提升拟合能力;针对定位精度的问题,利用数据增强方式扩充数据集,并在全连接层中增加 dropout 层;针对未知位置识别的问题,将 DCNN 与 I-Openmax(ImprovedOpenmax)算法相结合,增强未知位置识别能力。 实验结果表明,I-DCNN 的拟合能力明显优于 DCNN 和 AlexNet 模型,在包含未知位置的测试集上准确率为 96%,均方误差为 0. 061,平均绝对误差为 0. 054,性能明显优于基于支持向量回归(Support Vector Regression,SVR)定位算法。
Abstract:
In indoor positioning,it is a new method to use Deep Convolutional Neural Network (DCNN) to construct images and implement location classification for Channel State Information (CSI). However,DCNN-based indoor positioning algorithm is affected by theactual environment,data collection methods,and model structure,and has the problems of weak fitting ability,poor positioning accuracy,and weak recognition of unknown positions. It is difficult to meet the needs for positioning accuracy and stability in actual positioning.To solve these problems,an Improved Deep Convolutional Neural Network (I-DCNN) model that can identify unknown locations is proposed. The model structure of DCNN is adjusted to make it more complex to overcome under-fitting. And then,the data set is expandedthrough data enhancement and the dropout layer is added to the fully connected layer to improve positioning accuracy. Furthermore, IOpenmax (Improved-Openmax) algorithm is introduced to realize unknown location recognition together with DCNN. The experimentalresults show that the fitting ability of I-DCNN is significantly better than that of DCNN and AlexNet models. The accuracy rate on thetest set containing unknown positions is 96%,the mean square error is 0. 061,and the average absolute error is 0. 054. The performanceis significantly better than that based on Support Vector Regression (SVR) positioning algorithm.

相似文献/References:

[1]刘君 吴建国 褚曦丹 朱丽进 李炜.Cricket室内定位系统的研究与改进[J].计算机技术与发展,2011,(05):206.
　LIU Jun,WU Jian-guo,CHU Xi-dan,et al.Research and Improvement of Cricket Indoor Location System[J].,2011,(08):206.
[2]玄建永,王京春,陆耿,等.缩微智能车室内定位系统研究[J].计算机技术与发展,2014,24(01):1.
　XUAN Jian-yong,WANG Jing-chun,LU Geng,et al.Research on Indoor Position System for Autonomous Micro-vehicle[J].,2014,24(08):1.
[3]王凡,彭勇.基于TDOA的室内超声波定位方法的改进[J].计算机技术与发展,2014,24(06):250.
　WANG Fan,PENG Yong.Improved Indoor Location Method for Ultrasonic System Based on TDOA[J].,2014,24(08):250.
[4]杜月林[],石欣然[],王克寒[]. 基于ZigBee技术的室内定位系统算法研究及实现[J].计算机技术与发展,2014,24(07):245.
　DU Yue-lin[],SHI Xin-ran[],WANG Ke-han[]. Research and Realization of Indoor Positioning System Algorithm Based on ZigBee[J].,2014,24(08):245.
[5]玲玉,杨恒新,张昀. 基于无源UHF RFID的一种室内定位算法[J].计算机技术与发展,2015,25(06):110.
　QU Ling-yu,YANG Heng-xin,ZHANG Yun. An Indoor Location Algorithm Based on Passive UHF RFID[J].,2015,25(08):110.
[6]崔凤焦.表情识别算法研究进展与性能比较[J].计算机技术与发展,2018,28(02):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
　CUI Feng-jiao.Ｒesearch and Performance Comparison of Facial Expression Ｒecognition Algorithm[J].,2018,28(08):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
[7]王语琪[][],巩应奎[]. 一种基于视觉信息的可见光通信室内定位方法[J].计算机技术与发展,2016,26(01):200.
　WANG Yu-qi[][],GONG Ying-kui[]. An Indoor Positioning Method of Visible Light Communication Based on Visual Information[J].,2016,26(08):200.
[8]张丹丹,李雷. 基于PCANet-RF的人脸检测系统[J].计算机技术与发展,2016,26(02):31.
　ZHANG Dan-dan,LI Lei. Face Detection System Based on PCANet-RF[J].,2016,26(08):31.
[9]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].,2018,28(08):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[10]郭子琰,舒心,刘常燕,等.基于ReLU 函数的卷积神经网络的花卉识别算法[J].计算机技术与发展,2018,28(05):154.[doi:10．3969/j．issn．1673－629X．2018．05．035]
　GUO Ziyan,SHU Xin,LIU Changyan,et al.A Recognition Algorithm of Flower Based on Convolution Neural Network with ReLU Function[J].,2018,28(08):154.[doi:10．3969/j．issn．1673－629X．2018．05．035]
[11]尹铁源,刘 祺.基于卷积神经网络的室内定位方法的研究[J].计算机技术与发展,2021,31(增刊):31.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 006]
　YIN Tie-yuan,LIU Qi.Study on Indoor Localization Method Based on Convolutional Neural Network[J].,2021,31(08):31.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 006]

更新日期/Last Update: 2022-08-10