| 摘要: |
| 针对基于机器学习的可见光室内定位方法存在的手工调参、定位精度低等问题，结合蛇优化（Snake Optimization,SO）算法的寻优能力与卷积神经网络（Convolutional Neural Network,CNN）处理复杂非线性问题的能力，提出了一种基于SO-CNN模型的可见光室内定位优化方法。在考虑多径效应影响的情况下，采集每个位置点处的信噪比和对应位置坐标构建指纹数据库，对SO-CNN模型进行训练和测试，以得到最佳定位模型。实验结果表明，在5 m×5 m×3 m的房间中，与未经优化的CNN相比，该方法的平均定位误差降低了35.13%；与反向传播神经网络（Back Propagation Neural Network,BPNN）、多层感知器（Multilayer Perceptron,MLP）、SO-MLP相比，该方法的平均定位误差分别降低了54.75%，48.08%，37.01%。 |
| 关键词: 可见光室内定位(VLIP) 指纹定位法 蛇优化算法 卷积神经网络 |
| DOI：10.20079/j.issn.1001-893x.230616002 |
|
| 基金项目:江苏省“双创博士”项目（JSSCBS20211035）；江苏省博士后基金项目（2021K402C）；汽车新技术安徽省工程技术研究中心开放课题项目(QCKJ202205A) |
|
| An Optimization Method for Visible Light Indoor Positioning Based on SO-CNN |
| CHEN Jing,LIU Xuan,WANG Jinyuan,ZHANG Yonglong,ZHU Junwu |
| (1.College of Information EngineeringCollege of Artificial Intelligence,Yangzhou University,Yangzhou 225127,China;2.School of Computer Science and Engineering,Southeast University,Nanjing 211189,China;3.Automotive New Technique of Anhui Province Engineering Technology Research Center,Anhui Polytechnic University,Wuhu 241000,China) |
| Abstract: |
| To solve the problems of manual parameter adjustment and low positioning accuracy in machine learning-based visible light indoor positioning(VLIP),VLIP optimization approach based on the Snake Optimization(SO) algorithm and Convolutional Neural Network(CNN)(SO-CNN) model is proposed,which combines the optimization-seeking capabilities of the SO algorithm with the ability of the CNN to handle complex nonlinear problems.Considering the influence of multipath effects,the fingerprint database is constructed by collecting the signal-to-noise ratio at each location point and the corresponding coordinate,and the SO-CNN model is trained and tested to obtain the best positioning model.Experimental results show that in a room of 5 m×5 m×3 m,the average positioning error of the proposed method is reduced by 35.13% compared with that of an unoptimized CNN,and by 54.75%,48.08%,and 37.01%,respectively,compared with that of Back Propagation Neural Network(BPNN),Multilayer Perceptron(MLP),and SO-MLP. |
| Key words: visible light indoor positioning(VLIP) fingerprinting positioning snake optimization algorithm convolutional neural network |
