ABSTRACT
Deep learning (DL) has been used for wireless signal analysis in many applications, e.g., indoor localization. By collecting measurement data of wireless signals from the environment, DL models can be trained to accurately predict the change of signal characteristics. However, constructing high-quality DL training data from a real experiment environment is often labor-intensive and time-consuming, which is the biggest obstacle to applying the newest DL model to wireless network research. To address such issues, we present DeepWiSim, a ray-tracing-based wireless signal simulator that automates the DL process from data generation to model training and evaluation. The demonstration shows that DeepWiSim can efficiently generate high-quality simulated wireless signal measurement data and simultaneously train and evaluate the DL model.
- 2021. Next-generation simulation technology to accelerate the 5G journey. https://www.ericsson.com/en/blog/2021/4/5g-simulation-omniverse-platformGoogle Scholar
- 2022. iBwave. https://www.ibwave.com/ibwave-designGoogle Scholar
- 2022. PyTorch. https://pytorch.orgGoogle Scholar
- 2022. Tensorflow. https://www.tensorflow.orgGoogle Scholar
- 2022. Wireless InSite. https://www.remcom.comGoogle Scholar
- Moustafa Abbas, Moustafa Elhamshary, Hamada Rizk, Marwan Torki, and Moustafa Youssef. 2019. WiDeep: WiFi-based Accurate and Robust Indoor Localization System using Deep Learning. In PerCom. 1--10.Google Scholar
- Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Sanatan Sharma, Abhishek Rajkumar Sethi, Deepak Vasisht, and Dinesh Bharadia. 2020. Deep Learning Based Wireless Localization for Indoor Navigation. In MobiCom. 1--14.Google Scholar
- Alexey Dosovitskiy, German Ros, Felipe Codevilla, Antonio Lopez, and Vladlen Koltun. 2017. CARLA: An Open Urban Driving Simulator. In Proceedings of the 1st Annual Conference on Robot Learning. 1--16.Google Scholar
- Epic Games. 2022. Unreal Engine. https://www.unrealengine.comGoogle Scholar
- Xiansheng Guo, Nirwan Ansari, Fangzi Hu, Yuan Shao, Nkrow Raphael Elikplim, and Lin Li. 2020. A Survey on Fusion-Based Indoor Positioning. IEEE Communications Surveys & Tutorials 22, 1 (2020), 566--594.Google ScholarDigital Library
- Richard Kurle, Syama Sundar Rangapuram, Emmanuel de Bézenac, Stephan Günnemann, and Jan Gasthaus. 2020. Deep rao-blackwellised particle filters for time series forecasting. NeurIPS 33 (2020), 15371--15382.Google Scholar
- Longyuan Li, Jian Yao, Li Wenliang, Tong He, Tianjun Xiao, Junchi Yan, David Wipf, and Zheng Zhang. 2021. GRIN: Generative Relation and Intention Network for Multi-agent Trajectory Prediction. NeurIPS 34 (2021).Google Scholar
- Syama Sundar Rangapuram, Matthias W Seeger, Jan Gasthaus, Lorenzo Stella, Yuyang Wang, and Tim Januschowski. 2018. Deep state space models for time series forecasting. NeurIPS 31 (2018).Google Scholar
- David Salinas, Valentin Flunkert, Jan Gasthaus, and Tim Januschowski. 2020. DeepAR: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting 36, 3 (2020), 1181--1191.Google ScholarCross Ref
- Shital Shah, Debadeepta Dey, Chris Lovett, and Ashish Kapoor. 2017. AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles. In Field and Service Robotics. arXiv:arXiv:1705.05065 https://arxiv.org/abs/1705.05065Google Scholar
- Bohan Tang, Yiqi Zhong, Ulrich Neumann, Gang Wang, Siheng Chen, and Ya Zhang. 2021. Collaborative uncertainty in multi-agent trajectory forecasting. NeurIPS 34 (2021).Google Scholar
Index Terms
- DeepWiSim: a wireless signal simulator for automatic deep learning
Recommendations
Visual Data Simulation for Deep Learning in Robot Manipulation Tasks
Modelling and Simulation for Autonomous SystemsAbstractThis paper introduces the usage of simulated images for training convolutional neural networks for object recognition and localization in the task of random bin picking. For machine learning applications, a limited amount of real world image data ...
Analysis and modeling GPS NLOS effect in highly urbanized area
Current GPS positioning accuracy in urban areas is still unsatisfactory for various applications, including pedestrian navigation and autonomous driving. Due to the ineffectiveness of special corrector designs against non-line-of-sight (NLOS) reception, ...
A Comparative Study of Localization Methods in Indoor Environments
A comparative study, based on three different measurements (direction of ray arrival, time difference of arrival and received signal strength), to compute the unknown position of mobile stations in indoor environments is presented in this paper. The ...
Comments