arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.06394cs.LGeess.SPstat.ML

用于单接入点CSI室内定位的CNN-LSTM量化器

A CNN-LSTM Quantifier for Single Access Point CSI Indoor Localization

Minh Tu Hoang, Brosnan Yuen, Kai Ren, Xiaodai Dong, Tao Lu, Hung Le Nguyen, Robert Westendorp, Kishore Reddy

备注：
Channel state information (CSI), WiFi indoor localization, convolutional neural network, long short-term memory, fingerprint-based localization
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出CNN-LSTM量化网络用于单接入点WiFi CSI指纹室内定位，同时提取时空特征并以量化模型替代传统分类模型，实验表明平均定位误差2.5米，80%误差低于4米，性能优于其他算法约50%。

摘要：

本文提出了一种卷积神经网络（CNN）与长短期记忆（LSTM）量化器相结合的联合网络结构，用于WiFi指纹室内定位。与仅利用空间数据和分类模型的传统方法相比，我们的CNN-LSTM网络从单个路由器中提取接收信道状态信息（CSI）的空间和时间特征。此外，所提出的网络构建的是一个量化模型，而非大多数文献工作中采用的有限分类模型，这使得能够估计与参考点不完全相同的测试点。我们分析了CSI的不稳定性，并展示了一种使用综合滤波和归一化方案的缓解方法。通过在数百个测试位置上使用包括手机（Nexus 5）和笔记本电脑（Intel 5300 NIC）在内的多种移动设备进行大量现场实验，研究了定位精度。仅使用单个WiFi路由器，我们的结构实现了2.5米的平均定位误差，其中80%的误差在4米以内，在相同测试环境下优于其他已报道算法约50%。

英文摘要：

This paper proposes a combined network structure between convolutional neural network (CNN) and long-short term memory (LSTM) quantifier for WiFi fingerprinting indoor localization. In contrast to conventional methods that utilize only spatial data with classification models, our CNN-LSTM network extracts both space and time features of the received channel state information (CSI) from a single router. Furthermore, the proposed network builds a quantification model rather than a limited classification model as in most of the literature work, which enables the estimation of testing points that are not identical to the reference points. We analyze the instability of CSI and demonstrate a mitigation solution using a comprehensive filter and normalization scheme. The localization accuracy is investigated through extensive on-site experiments with several mobile devices including mobile phone (Nexus 5) and laptop (Intel 5300 NIC) on hundreds of testing locations. Using only a single WiFi router, our structure achieves an average localization error of 2.5~m with $\mathrm{80\%}$ of the errors under 4~m, which outperforms the other reported algorithms by approximately $\mathrm{50\%}$ under the same test environment.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑