OmniCells: Cross-Device Cellular-based Indoor Location Tracking Using Deep Neural Networks

The demand for a ubiquitous and accurate indoor localization service is continuously growing. Cellular-based systems are a good candidate to provide such ubiquitous service due to their wide availability worldwide. One of the main barriers for accuracy is the large number of models of cell phones, which results in variations of the measured received signal strength (RSS), even at the same location and time. In this paper, we propose OmniCells, a deep learning-based system that leverages cellular measurements from one or more training devices to provide a consistent performance across unseen tracking phones. Specifically, OmniCells uses a novel approach to multi-task learning based on autoencoders that allows it to learn a rich and device-invariant RSS representation without any assumptions about the source or target devices. OmniCells also incorporates different modules to boost the system’s accuracy with RSS relative difference-based features and to improve the deep model’s generalization and robustness. Evaluation of OmniCells in two realistic testbeds using different Android phones with different form factors and cellular radio hardware shows that OmniCells can achieve a consistent median localization accuracy when tested on different phones. This is better than the state-of-the-art indoor cellular-based systems by at least 101%.

[1]  Moustafa Youssef,et al.  WiDeep: WiFi-based Accurate and Robust Indoor Localization System using Deep Learning , 2019, 2019 IEEE International Conference on Pervasive Computing and Communications (PerCom.

[2]  Hamada Rizk.  Device-Invariant Cellular-Based Indoor Localization System Using Deep Learning , 2019 .

[3]  Yoshua Bengio,et al.  Practical Recommendations for Gradient-Based Training of Deep Architectures , 2012, Neural Networks: Tricks of the Trade.

[4]  Xiangyu Wang,et al.  ResLoc: Deep residual sharing learning for indoor localization with CSI tensors , 2017, 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).

[5]  Eyal de Lara,et al.  GSM indoor localization , 2007, Pervasive Mob. Comput..

[6]  Lin Ma,et al.  Linear Regression Algorithm against Device Diversity for Indoor WLAN Localization System , 2017, GLOBECOM 2017 - 2017 IEEE Global Communications Conference.

[7]  Fangfang Dong,et al.  A Calibration-Free Localization Solution for Handling Signal Strength Variance , 2009, MELT.

[8]  Ramesh Govindan,et al.  Energy-efficient positioning for smartphones using Cell-ID sequence matching , 2011, MobiSys '11.

[9]  Jason Jianjun Gu,et al.  Deep Neural Networks for wireless localization in indoor and outdoor environments , 2016, Neurocomputing.

[10]  Mohamed Ibrahim,et al.  Enabling wide deployment of GSM localization over heterogeneous phones , 2013, 2013 IEEE International Conference on Communications (ICC).

[11]  Khaled A. Harras,et al.  WiNar: RTT-based Sub-meter Indoor Localization using Commercial Devices , 2020, 2020 IEEE International Conference on Pervasive Computing and Communications (PerCom).

[12]  Mohamed Ibrahim,et al.  CellSense: An Accurate Energy-Efficient GSM Positioning System , 2011, IEEE Transactions on Vehicular Technology.

[13]  Moustafa Youssef,et al.  MonoDCell: A Ubiquitous and Low-Overhead Deep Learning-based Indoor Localization with Limited Cellular Information , 2019, SIGSPATIAL/GIS.

[14]  Shiwen Mao,et al.  PhaseFi: Phase Fingerprinting for Indoor Localization with a Deep Learning Approach , 2014, 2015 IEEE Global Communications Conference (GLOBECOM).

[15]  Xiangyu Wang,et al.  CiFi: Deep convolutional neural networks for indoor localization with 5 GHz Wi-Fi , 2017, 2017 IEEE International Conference on Communications (ICC).

[16]  Christoforos Panayiotou,et al.  Localization Using Radial Basis Function Networks and Signal Strength Fingerprints in WLAN , 2009, GLOBECOM 2009 - 2009 IEEE Global Telecommunications Conference.

[17]  Moustafa Youssef,et al.  Effectiveness of Data Augmentation in Cellular-based Localization Using Deep Learning , 2019, 2019 IEEE Wireless Communications and Networking Conference (WCNC).

[18]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[19]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[20]  Shiwen Mao,et al.  BiLoc: Bi-Modal Deep Learning for Indoor Localization With Commodity 5GHz WiFi , 2017, IEEE Access.

[21]  E. B. Andersen,et al.  Information Science and Statistics , 1986 .

[22]  Mohamed Ibrahim,et al.  A Hidden Markov Model for Localization Using Low-End GSM Cell Phones , 2011, 2011 IEEE International Conference on Communications (ICC).

[23]  Moustafa Youssef,et al.  Crescendo: An Infrastructure-free Ubiquitous Cellular Network-based Localization System , 2019, 2019 IEEE Wireless Communications and Networking Conference (WCNC).

[24]  Moustafa Youssef,et al.  CellinDeep: Robust and Accurate Cellular-Based Indoor Localization via Deep Learning , 2019, IEEE Sensors Journal.

[25]  Shiwen Mao,et al.  DeepFi: Deep learning for indoor fingerprinting using channel state information , 2015, 2015 IEEE Wireless Communications and Networking Conference (WCNC).

[26]  Sanghyuk Lee,et al.  A scalable deep neural network architecture for multi-building and multi-floor indoor localization based on Wi-Fi fingerprinting , 2017, ArXiv.

[27]  Moustafa Youssef,et al.  DeepLoc: a ubiquitous accurate and low-overhead outdoor cellular localization system , 2018, SIGSPATIAL/GIS.

[28]  Mikkel Baun Kjærgaard,et al.  Hyperbolic Location Fingerprinting: A Calibration-Free Solution for Handling Differences in Signal Strength (concise contribution) , 2008, 2008 Sixth Annual IEEE International Conference on Pervasive Computing and Communications (PerCom).

[29]  Hamada Rizk.  SoloCell: Efficient Indoor Localization Based on Limited Cell Network Information And Minimal Fingerprinting , 2019, SIGSPATIAL/GIS.

[30]  Shih-Hau Fang,et al.  Calibration-Free Approaches for Robust Wi-Fi Positioning against Device Diversity: A Performance Comparison , 2012, 2012 IEEE 75th Vehicular Technology Conference (VTC Spring).

[31]  Amany Sarhan,et al.  A hybrid outlier detection algorithm based on partitioning clustering and density measures , 2015, 2015 Tenth International Conference on Computer Engineering & Systems (ICCES).

[32]  Weixiong Rao,et al.  Transfer Learning-Based Outdoor Position Recovery with Telco Data , 2019, ArXiv.

[33]  Mario Siller,et al.  A fingerprinting indoor localization algorithm based deep learning , 2016, 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN).

[34]  Klaus-Robert Müller,et al.  Efficient BackProp , 2012, Neural Networks: Tricks of the Trade.

[35]  Bruce Denby,et al.  Robust indoor localization and tracking using GSM fingerprints , 2015, EURASIP J. Wirel. Commun. Netw..

[36]  Shahrokh Valaee,et al.  Automatic Device-Transparent RSS-Based Indoor Localization , 2014, 2015 IEEE Global Communications Conference (GLOBECOM).

[37]  Keith W. Ross,et al.  Computer networking - a top-down approach featuring the internet , 2000 .