I am a Ph.D. student at UCSD, advised by Prof. Dinesh Bharadia and co-advised by Prof. Deepak Vasisht. My research interests spans over the areas of deep learning and wireless sensing, with applications to the IoT's, mobile computing, indoor navigation, and robotics. My research vision is to make everyday things intelligent and connected so that people can live in an unprecedentedly smart environment.

Projects

cm-Accurate, Real time, and scalable UWB based Indoor 3D Localization

webpage

Since UWB has been developed as loclaization specific protocol, there has been a need for infrastructure based, low-power and real-time indoor localization while providing cm-Accurate 3D UWB tag locations. We solve these problems by novel hardware, firmware and algorithm designs. Where wedesigning our custom UWB anchor hardware and firmware that enables accurate 3D AoA, which we utilize to get cm-accurate 3D location using our novel algorithms.

  • ULoc: Low-Power, Scalable and cm-Accurate UWB-Tag Localizationand Tracking for Indoor Applications
    Minghui Zhao, Tyler Chang, Aditya Arun, Roshan Ayyalasomayajula, Chi Zhang, Dinesh Bharadia
    IMWUT, 2021
    Presented at Ubicomp, 2021
    [paper]-[ppt]-[20-min video]-[6-min video]-[Demo 1]-[Demo 2]-[Source]

Deep-Learning and Context assisted Indoor Wireless localization

webpage

There has been a lot of work in Indoor WiFi localization in the past decade, with none of them being deployed in real-life. With this project we intend to bridge the gap between the real world maps and the WiFi maps and enable deep-learning based solution by allowing large scale data-collection.

  • Sound source localization based on multi-task learning and image translation network
    Yifan Wu, Roshan Ayyalasomayajula, Michael J. Bianco, Dinesh Bharadia, Peter Gerstoft
    JASA, Nov 2021
    [paper]

  • SSLIDE: Blind Sound Source Localization based on Deep Learning
    Yifan Wu, Roshan Ayyalasomayajula, Michael J. Bianco, Dinesh Bharadia, Peter Gerstoft
    ICASSP, Jun 2021
    [paper]-[ppt]-[video]

  • OpenSourcing: Wireless Indoor Localization Datasets (WILD)
    [Datasets]
    Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Dinesh Bharadia
  • Deep Learning based Wireless Localization for Indoor Navigation
    Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Sanatan Sharma, Abhishek Sethi Deepak Vasisht, Dinesh Bharadia
    ACM Mobicom, 2020
    [paper]-[ppt]-[video]-[codes]-[Datasets]

Accurate Wi-Fi Anchor Localization

webpage

With the advent of CSI based WiFi localization, the indoor localization paradigm has come down to decimenter level accurate location estiamtes. While this remains the case, the analysis and solution on how important the accurate anchor location is unresolved which is solved in this work.



  • LocAP: Accurate Localization of Existing WiFi Infrastructure
    Roshan Ayyalasomayajula, Aditya Arun, Chenfeng Wu, Shrivatsan Rajagopalan, Shreya Ganesaraman, Aravind Seetharaman, Dinesh Bharadia
    USENIX NSDI, 2020
    [paper]-[ppt]-[video]

BLE based localization

webpage

State of the art localization using Bluetooth Low Energy (BLE) devices is baed on RSSI, while Wi-Fi has achieved decimeter level Localization using Channel State Information (CSI) based localization. This work brings CSI based sub-meter level localization to BLE tags.


Update:BLoc, published in December 2018 has propsoed the idea of long sequences of 1's and 0's and BLE 5.1, released in January 2019, which uses CTE for AoA estimation that is similar to BLoc's implementation.

  • BLoc: CSI-based accurate localization for BLE tags
    Roshan Ayyalasomayajula, Deepak Vasisht, Dinesh Bharadia
    ACM CoNEXT, 2018
    [paper]-[pptx]-[video]