publications

2026

1. ACM MobiHoc ’26

   IoTell: Multimodal Sensor Fusion for Enhanced IoT Control

   Md Fazlay Masum Rabbi Billah, Amod K. Agrawal, Viswajith Govinda Rajan, and 1 more author

   In Proceedings of the 27th International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc ’26), Tokyo, Japan, Nov 2026

   Accepted August 23, 2026

   Abs Bib

   Current control interfaces between a user and a network of heterogeneous IoT devices largely consist of device-specific applications or voice commands with a fixed set of instructions for each device. Such control mechanisms fail to cope with the vision of dense trillion-node IoT network, where users would interact with tens of smart devices in everyday environments. Recent advances in RF-based ranging technologies enable accurate estimation of user location in indoor environments. However, leveraging this spatial context to enable intuitive interaction remains an open challenge, particularly in dense deployments where multiple candidate devices exist within close proximity. In this paper, we introduce IoTell, an intuitive control interface that allows users to interact with a device by pointing to it and issuing simple voice commands. IoTell fuses location, orientation, and voice instructions to identify the intended device without requiring explicit device naming. Operating in device-dependent mode, the system estimates user location and orientation through a wearable device using RF-based localization and inertial sensing. In the device-independent mode, IoTell enables the user to have a similar experience even when the user may not carry a wearable, by estimating user location and orientation through multimodal fusion of audio and RF signal features. We implement IoTell using a UWB-based localization system and inertial sensing on a wearable device, and evaluate its performance in a multi-device indoor environment. Our experiments demonstrate that IoTell achieves 95.2% accuracy in correctly executing commands in device-dependent settings and 88.3% accuracy, with a mean localization error of 45 cm, in device-independent scenarios. These results show that multimodal fusion of spatial and linguistic context can significantly improve interaction with IoT systems under realistic conditions.

   @inproceedings{billah2026iotell,
     author = {Billah, Md Fazlay Masum Rabbi and Agrawal, Amod K. and Rajan, Viswajith Govinda and Campbell, Brad},
     title = {IoTell: Multimodal Sensor Fusion for Enhanced IoT Control},
     booktitle = {Proceedings of the 27th International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc '26)},
     month = nov,
     year = {2026},
     location = {Tokyo, Japan},
     publisher = {ACM},
     address = {New York, NY, USA},
     note = {Accepted August 23, 2026}
   }

2. ACM HumanSys (MobiCom ’26)

   HeadSense: Head Orientation Tracking for AI Wearables using Ultra-Wideband Ranging

   Amod K. Agrawal, Ashutosh Dhekne, and Romit Roy Choudhury

   In Proceedings of the International Workshop on Human-Centered Sensing, Modeling, and Intelligent Systems (HumanSys ’26), co-located with ACM MobiCom 2026, Austin, TX, USA, Aug 2026

   Accepted August 14, 2026

   Abs Bib

   AI wearables (smart glasses, pins, pendants, and earbuds) use head gestures for interaction, yet lack a reliable, continuous measure of head-to-torso orientation. Such a measurement would enable temporal attention context for AI assistants, allowing them to infer sustained visual attention, disambiguate spatial commands, and filter out incidental head movements. IMU-based tracking drifts over time, limiting it to transient gesture detection such as nods and shakes. Magnetometer-based yaw correction fails indoors near ferrous materials and speaker magnets common in wearable devices. We present HeadSense, a necklace-like wearable that uses ultra-wideband (UWB) radios to continuously track head orientation relative to the torso with bounded, non-accumulating error. HeadSense introduces a wired reference path between chest-mounted transmitters and head-worn receivers, enabling relative time-of-flight measurements with sub-centimeter precision. A four-stage signal processing pipeline combines channel impulse response (CIR) interpolation, template matching, phase refinement, and IMU-based ambiguity resolution to reduce the typical 10 cm UWB ranging error to a median of 5 mm and achieve a median orientation error of 6.5°, validated against Vicon motion capture ground truth. HeadSense provides an infrastructure-free, body-centric orientation sensing capability for AI wearables without requiring cameras or external tracking systems.

   @inproceedings{agrawal2026headsense,
     author = {Agrawal, Amod K. and Dhekne, Ashutosh and Choudhury, Romit Roy},
     title = {HeadSense: Head Orientation Tracking for AI Wearables using Ultra-Wideband Ranging},
     booktitle = {Proceedings of the International Workshop on Human-Centered Sensing, Modeling, and Intelligent Systems (HumanSys '26), co-located with ACM MobiCom 2026},
     month = aug,
     year = {2026},
     location = {Austin, TX, USA},
     publisher = {ACM},
     address = {New York, NY, USA},
     note = {Accepted August 14, 2026}
   }

3. ACM HumanSys (MobiCom ’26)

   Vision: HomeLM for Ambient Intelligence in the Smart Home

   Amod K. Agrawal

   In Proceedings of the International Workshop on Human-Centered Sensing, Modeling, and Intelligent Systems (HumanSys ’26), co-located with ACM MobiCom 2026, Austin, TX, USA, Aug 2026

   Accepted August 14, 2026

   Abs Bib

   Smart homes hold abundant sensing opportunities, from microphone arrays and presence radars to wearables and everyday device interactions. These modalities rarely add up to a coherent picture of what is happening inside the home. Today’s ambient intelligence stitches together many narrow, task-specific models but lacks a unified semantic understanding of the signals they consume. We envision HomeLM, a foundation model that learns to align home-sensing signals with natural language, so that a single model can describe and reason about human activity across a whole house rather than classifying it one task at a time. This would let a home recognize situations it was never explicitly trained for and act proactively on behalf of its occupants. This paper outlines our vision for HomeLM: how such a model could be trained, the research challenges in realizing it, and the opportunities it brings for the smart home.

   @inproceedings{agrawal2026homelm,
     author = {Agrawal, Amod K.},
     title = {Vision: HomeLM for Ambient Intelligence in the Smart Home},
     booktitle = {Proceedings of the International Workshop on Human-Centered Sensing, Modeling, and Intelligent Systems (HumanSys '26), co-located with ACM MobiCom 2026},
     month = aug,
     year = {2026},
     location = {Austin, TX, USA},
     publisher = {ACM},
     address = {New York, NY, USA},
     note = {Accepted August 14, 2026}
   }

4. U.S. Patent

   Multi-Modal Indoor Mapping and Device Localization System

   Amod K. Agrawal, Abhishek Bafna, and Ankit Shah

   Patent Application, United States Patent and Trademark Office, Jun 2026

   U.S. Patent App. 19/718,581, Pending

   Abs Bib

   This patent application discloses a multi-modal system for indoor mapping and device localization. By fusing signals across multiple sensing modalities, the system constructs a spatial map of an indoor environment and localizes smart devices within it. The approach enables accurate, context-aware placement of devices on the inferred map, supporting spatially aware automation and richer ambient experiences without requiring manual configuration.

   @patent{agrawal2026maplocalization,
     author = {Agrawal, Amod K. and Bafna, Abhishek and Shah, Ankit},
     title = {Multi-Modal Indoor Mapping and Device Localization System},
     month = jun,
     year = {2026},
     school = {Patent Application},
     location = {United States Patent and Trademark Office},
     number = {19/718,581},
     note = {U.S. Patent App. 19/718,581, Pending},
     address = {United States}
   }

2025

1. ACM EnvSys (MobiSys ’25)

   Localization using Angle-of-Arrival Triangulation

   Amod K. Agrawal

   In Proceedings of the International Workshop on Environmental Sensing Systems for Smart Cities (EnvSys ’25), co-located with MobiSys 2025, Anaheim, CA, USA, Jun 2025

   Abs DOI arXiv Bib Slides

   Indoor localization is a long-standing challenge in mobile computing, with significant implications for enabling location-aware and intelligent applications within smart environments such as homes, offices, and retail spaces. As AI assistants such as Amazon Alexa and Google Nest become increasingly pervasive, microphone-equipped devices are emerging as key components of everyday life and home automation. This paper introduces a passive, infrastructure-light system for localizing human speakers using speech signals captured by two or more spatially distributed smart devices. The proposed approach, GCC+, extends the Generalized Cross-Correlation with Phase Transform (GCC-PHAT) method to estimate the Angle-of-Arrival (AoA) of audio signals at each device and applies robust triangulation techniques to infer the speaker’s two-dimensional position. To further improve temporal resolution and localization accuracy, feature-space expansion and subsample interpolation techniques are employed for precise Time Difference of Arrival (TDoA) estimation. The system operates without requiring hardware modifications, prior calibration, explicit user cooperation, or knowledge of the speaker’s signal content, thereby offering a highly practical solution for real-world deployment. Experimental evaluation in a real-world home environment yields a median AoA estimation error of 2.2 degree and a median localization error of 1.25m, demonstrating the feasibility and effectiveness of audio-based localization for enabling context-aware, privacy-preserving ambient intelligence.

   @inproceedings{agrawal2025aoa,
     author = {Agrawal, Amod K.},
     title = {Localization using Angle-of-Arrival Triangulation},
     booktitle = {Proceedings of the International Workshop on Environmental Sensing Systems for Smart Cities (EnvSys '25), co-located with MobiSys 2025},
     month = jun,
     year = {2025},
     location = {Anaheim, CA, USA},
     pages = {1--6},
     publisher = {ACM},
     address = {New York, NY, USA},
     doi = {10.1145/3742460.3742984},
     isbn = {979-8-4007-1986-8},
   }

2. U.S. Patent

   Localization of devices in a structure

   Amod K. Agrawal, Chaitanya Desai, Lakshmi Venkatraman, and 3 more authors

   Amazon Technologies, Inc., Seattle, WA, United States Patent and Trademark Office, Aug 2025

   U.S. Patent App. 18/542252, Granted 08/12/2025

   Abs URL Bib

   Disclosed are various embodiments for localizing smartdevices in a structure. An image of a portion of at least one room of a structure may be rendered on a display of the mobile device. A point in the image may be determined where a nonlocalized device is depicted, the nonlocalized device being positioned in the at least one room of the structure. The nonlocalized device is localized in the at least one room of the structure based on a position and orientation of the mobile device relative to the at least one room of the structure and based on a location of the point in the image. Alternatively, a location of a nonlocalized device in the structure may be determined by using the location of a mobile device as proxy.

   @patent{localizationinastructure,
     author = {Agrawal, Amod K. and Desai, Chaitanya and Venkatraman, Lakshmi and Bocca, Maurizio and Argyropoulos, Paraskevas and Kim, Wontak},
     title = {Localization of devices in a structure},
     month = aug,
     year = {2025},
     school = {Amazon Technologies, Inc., Seattle, WA},
     location = {United States Patent and Trademark Office},
     number = {US-12387374-B1},
     note = {U.S. Patent App. 18/542252, Granted 08/12/2025},
     url = {https://patents.google.com/patent/US12387374B1/en},
     address = {United States}
   }

3. IoT World Today

   Ultra-Wideband: The Next Innovation in Location-Based Services

   Amod K. Agrawal

   IoT World Today, Jun 2025

   Published June 17, 2025; Contributing author

   Abs URL

   Ultra-Wideband (UWB) technology is rapidly emerging as a key enabler of high-precision location-based services across industries. This article explores how UWB’s fine-grained spatial accuracy, low latency, and robustness to interference make it ideal for applications such as indoor navigation, asset tracking, and secure device-to-device communication. Unlike traditional wireless technologies like Wi-Fi or Bluetooth, UWB offers centimeter-level accuracy, which opens new possibilities in smart homes, logistics, automotive, and augmented reality. The piece also highlights recent adoption trends, standardization efforts, and future directions for UWB in the broader context of IoT connectivity and context-aware experiences.

   @article{iotworldtoday2025uwb,
     author = {Agrawal, Amod K.},
     title = {Ultra-Wideband: The Next Innovation in Location-Based Services},
     howpublished = {\url{https://www.iotworldtoday.com/connectivity/ultra-wideband-the-next-innovation-in-location-based-services}},
     journal = {IoT World Today},
     month = jun,
     year = {2025},
     url = {https://www.iotworldtoday.com/connectivity/ultra-wideband-the-next-innovation-in-location-based-services},
     note = {Published June 17, 2025; Contributing author}
   }

4. IEEE Computer Society

   Can Sensing Be Safe? Designing Privacy-Aware Wireless Systems

   Amod K. Agrawal

   IEEE Computer Society, Jul 2025

   Published July 1, 2025; Publication under IEEE computer.org

   Abs URL

   From fitness trackers to smart speakers, mobile sensing has quietly become ubiquitous, embedding itself into our daily lives. These devices are capable of monitoring motion, detecting presence, identifying user activities, and even inferring health conditions, sometimes without any user interaction. As these applications grow and become central to smart environments, a critical question arises: Can sensing systems be designed to be both safe and ethical?

   @article{ieeecs2025safesensing,
     author = {Agrawal, Amod K.},
     title = {Can Sensing Be Safe? Designing Privacy-Aware Wireless Systems},
     journal = {IEEE Computer Society},
     year = {2025},
     month = jul,
     howpublished = {\url{https://www.computer.org/publications/tech-news/trends/safe-sensing}},
     url = {https://www.computer.org/publications/tech-news/trends/safe-sensing},
     note = {Published July 1, 2025; Publication under IEEE computer.org}
   }

5. IoT For All

   Bluetooth Channel Sounding: A Quiet Revolution in Indoor Sensing

   Amod K. Agrawal

   IoT For All, Jul 2025

   Published July 2, 2025; Contributing author

   Abs URL

   As AI, smart environments, and ambient computing continue to evolve, Bluetooth Channel Sounding offers a simple but powerful promise: spatial intelligence without the hardware deployment complexity that has traditionally slowed widespread adoption.

   @article{agrawal2025bluetooth,
     author = {Agrawal, Amod K.},
     title = {Bluetooth Channel Sounding: A Quiet Revolution in Indoor Sensing},
     journal = {IoT For All},
     year = {2025},
     month = jul,
     howpublished = {\url{https://www.iotforall.com/bluetooth-channel-sounding-indoor-sensing}},
     url = {https://www.iotforall.com/bluetooth-channel-sounding-indoor-sensing},
     note = {Published July 2, 2025; Contributing author}
   }

6. IoT For All

   From Warning Lights to Insights: Vehicular IoT and Predictive Maintenance

   Amod K. Agrawal

   IoT For All, Jul 2025

   Published July 18, 2025; Contributing author

   Abs URL

   Modern vehicles are packed with technology, yet for most drivers, the experience of car maintenance remains frustratingly reactive. A warning light typically turns on, often cryptic and sometimes urgent, prompting consumers to diagnose the issue themselves or schedule a repair. While today’s vehicles contain hundreds of embedded sensors generating continuous performance data, much of that information remains untapped for everyday drivers. What if cars could use their own data streams to anticipate problems, guide drivers in plain language, and even integrate with personal digital assistants? Recent advancements in edge computing, AI, and IoT connectivity are poised to make this vision a reality. Vehicles are quietly transforming into intelligent sensing platforms that don’t just react to faults but predict them. This helps drivers avoid unexpected breakdowns and turn the vehicle maintenance experience into a proactive one.

   @article{agrawal2025vehiculariot,
     author = {Agrawal, Amod K.},
     title = {From Warning Lights to Insights: Vehicular IoT and Predictive Maintenance},
     journal = {IoT For All},
     year = {2025},
     month = jul,
     howpublished = {\url{https://www.iotforall.com/ai-iot-car-technology-maintenance}},
     url = {https://www.iotforall.com/ai-iot-car-technology-maintenance},
     note = {Published July 18, 2025; Contributing author}
   }

7. Network World

   Ambient sensing: Privacy-aware embedded intelligence

   Amod K. Agrawal

   Network World, Jul 2025

   Published July 24, 2025; Contributing author

   Abs URL

   With the rise of next-gen wearables and smart home devices, AI systems are beginning to evolve beyond our screens and into the spaces around us. Across the industry, there’s a push to embed intelligence into the physical environments where we live and work. Ambient intelligence is emerging as a core concept in smart homes, enabling everyday devices to passively observe their surroundings and understand where people are, what they’re doing and what they might need — all without explicit user interaction.

   @article{agrawal2025passivesensing,
     author = {Agrawal, Amod K.},
     title = {Ambient sensing: Privacy-aware embedded intelligence},
     journal = {Network World},
     year = {2025},
     month = jul,
     howpublished = {\url{https://www.networkworld.com/article/4027389/ambient-sensing-privacy-aware-embedded-intelligence.html}},
     url = {https://www.networkworld.com/article/4027389/ambient-sensing-privacy-aware-embedded-intelligence.html},
     note = {Published July 24, 2025; Contributing author}
   }

8. IoT For All

   Enabling Ambient AI with Wearables and AIoT

   Amod K. Agrawal

   IoT For All, Sep 2025

   Published Sept 1, 2025; Contributing author

   Abs URL

   Artificial intelligence is beginning to move beyond text and screen-based systems into devices that sense and respond to the physical world. Wearable technologies such as smart glasses, earbuds, and watches can provide continuous information about the user, while sensors in the environment can detect presence and activity. Together, these systems create opportunities for AI that is proactive, context-aware, and able to support daily life in real time. This article examines how wearables and ambient sensing can work together, the challenges of processing data locally versus in the cloud, and the importance of addressing privacy and trust. The convergence of wearables, ubiquitous sensing, and agentic AI marks a critical step toward realizing the vision of ambient AI.

   @article{agrawal2025ambientai,
     author = {Agrawal, Amod K.},
     title = {Enabling Ambient AI with Wearables and AIoT},
     journal = {IoT For All},
     year = {2025},
     month = sep,
     howpublished = {\url{https://www.iotforall.com/ambient-ai-aiot-wearables}},
     url = {https://www.iotforall.com/ambient-ai-aiot-wearables},
     note = {Published Sept 1, 2025; Contributing author}
   }

9. Network World

   HomeLM: A Foundation Model for Ambient AI in Smart Homes

   Amod K. Agrawal

   Network World, Sep 2025

   Published Sept 11, 2025; Contributing author

   Abs URL

   Smart homes today contain many connected devices, but these systems remain fragmented and limited to isolated functions. Achieving true Ambient AI requires moving beyond simple automation toward environments that can sense, interpret, and respond to human activity in a meaningful way. This article introduces HomeLM, a foundation model designed to unify diverse ambient sensor streams (such as BLE, UWB, Wi-Fi CSI, ultrasound, and mmWave) by aligning them with natural language. HomeLM can transform raw signals into human-understandable narratives of daily life, enabling proactive and context-aware intelligence. Drawing inspiration from models like CLIP and SensorLM, the paper discusses how hierarchical captioning and multimodal fusion can allow HomeLM to generalize across tasks, adapt with limited data, and provide interpretable explanations of sensor input. The article highlights the challenges of data availability, heterogeneity, privacy, and efficiency, and argues that addressing these will be key to realizing intelligent smart home environments.

   @article{agrawal2025homelm,
     author = {Agrawal, Amod K.},
     title = {HomeLM: A Foundation Model for Ambient AI in Smart Homes},
     journal = {Network World},
     year = {2025},
     month = sep,
     howpublished = {https://www.networkworld.com/article/4054975/homelm-a-foundation-model-for-ambient-ai.html},
     url = {https://www.networkworld.com/article/4054975/homelm-a-foundation-model-for-ambient-ai.html},
     note = {Published Sept 11, 2025; Contributing author}
   }

10. IEEE Computer

    Our Cars Can Talk: How IoT Brings AI to Vehicles

    Amod K. Agrawal

    Humanity and Computing, IEEE Computer, Magazine, Vol. 58, Issue 11, Nov 2025

    Editors: Dr. Jeffrey Voas and Prof. Domenico Talia

    Abs DOI arXiv Bib

    Bringing AI to vehicles and enabling them as sensing platforms is key to transforming maintenance from reactive to proactive. Now is the time to integrate AI copilots that speak both languages: machine and driver. This article offers a conceptual and technical perspective intended to spark interdisciplinary dialogue and guide future research and development in intelligent vehicle systems, predictive maintenance, and AI-powered user interaction.

    @article{agrawal2025CarsCanTalk,
      author = {Agrawal, Amod K.},
      title = {Our Cars Can Talk: How IoT Brings AI to Vehicles},
      journal = {Humanity and Computing, IEEE Computer},
      month = nov,
      year = {2025},
      location = {Magazine, Vol. 58, Issue 11},
      doi = {10.1109/MC.2025.3591918},
      pages = {8-11},
      publisher = {IEEE},
      address = {},
      note = {Editors: Dr. Jeffrey Voas and Prof. Domenico Talia}
    }

11. IoT For All

    Speech to Location: Enabling Indoor Localization with Microphones

    Amod K. Agrawal

    IoT For All, Oct 2025

    Published Oct 13, 2025; Contributing author

    Abs URL

    With millions of smart speakers already in homes, their built-in microphones offer a powerful way to enable indoor localization without new hardware. By combining Angle-of-Arrival estimates from multiple devices, user position can be triangulated in real time to drive presence-aware automation, personalized voice responses, and safety features — all while processing audio locally for privacy. This scalable approach turns today’s voice-enabled devices into the backbone of ambiently intelligent homes.

    @article{agrawal2025speechlocalization,
      author = {Agrawal, Amod K.},
      title = {Speech to Location: Enabling Indoor Localization with Microphones},
      journal = {IoT For All},
      year = {2025},
      month = oct,
      howpublished = {https://www.iotforall.com/indoor-positioning-localization-microphones},
      url = {https://www.iotforall.com/indoor-positioning-localization-microphones},
      note = {Published Oct 13, 2025; Contributing author}
    }

12. The AI Journal

    How AI Is Rewiring Smart Homes and Personal Computing

    Amod K. Agrawal

    The AI Journal, Nov 2025

    Published Nov 1, 2025; Contributing author

    Abs URL

    As smart speakers, TVs, wearables, and appliances fill our homes, AI is turning them into a distributed, intelligent system that learns patterns, predicts needs, and adapts living spaces in real time. By combining data from microphones, Wi-Fi, BLE, mmWave, and other sensors, AI can localize users, infer context, and coordinate multiple devices to act together — locking screens when you walk away, automating lighting and HVAC, or sending wellness alerts. Advances in on-device processing, federated learning, and Matter-based interoperability make these systems private, scalable, and responsive, pushing computing toward a future of seamless, ambient intelligence.

    @article{agrawal2025aiinsmarthomes,
      author = {Agrawal, Amod K.},
      title = {How AI Is Rewiring Smart Homes and Personal Computing},
      journal = {The AI Journal},
      year = {2025},
      month = nov,
      howpublished = {https://aijourn.com/how-ai-is-rewiring-smart-homes-and-personal-computing/},
      url = {https://aijourn.com/how-ai-is-rewiring-smart-homes-and-personal-computing/},
      note = {Published Nov 1, 2025; Contributing author}
    }

2024

1. U.S. Patent

   Methods for automatic device grouping from wireless signals

   Amod K. Agrawal, Maurizio Bocca, Md Fazlay Masum Rabbi Billah, and 2 more authors

   Patent Application, United States Patent and Trademark Office, Sep 2024

   U.S. Patent App. 18/900,343, Pending

   Abs Bib

   This patent application introduces techniques for automatically grouping devices based on observed wireless signals. The method enables context-aware behaviors, co-location detection, and spatial organization without requiring explicit configuration. The system leverages Bluetooth and Wi-Fi signal characteristics, signal strength, and transmission patterns to cluster devices and infer shared physical environments.

   @patent{agrawal2024devicegrouping,
     author = {Agrawal, Amod K. and Bocca, Maurizio and Billah, Md Fazlay Masum Rabbi and McClain, Sean and VijayaRaghavan, VenkatRaghavan},
     title = {Methods for automatic device grouping from wireless signals},
     month = sep,
     year = {2024},
     school = {Patent Application},
     location = {United States Patent and Trademark Office},
     note = {U.S. Patent App. 18/900,343, Pending},
     address = {United States}
   }

2023

1. U.S. Patent

   Methods and apparatus to localize devices in the home

   Amod K. Agrawal, Chaitanya Desai, Lakshmi Venkatraman, and 3 more authors

   Patent Application, United States Patent and Trademark Office, Dec 2023

   U.S. Patent App. 18/542,222, Pending

   Abs Bib

   This work describes methods and apparatus to determine the position of devices within a home environment using wireless technologies. By leveraging RF signals and signal processing techniques, the system enables accurate in-home localization for smart devices, enhancing contextual automation, device grouping, and user interaction.

   @patent{agrawal2024localization,
     author = {Agrawal, Amod K. and Desai, Chaitanya and Venkatraman, Lakshmi and Bocca, Maurizio and Argyropoulos, Paraskevas and Kim, Wontak},
     title = {Methods and apparatus to localize devices in the home},
     month = dec,
     year = {2023},
     school = {Patent Application},
     location = {United States Patent and Trademark Office},
     number = {18/542,222},
     note = {U.S. Patent App. 18/542,222, Pending},
     address = {United States}
   }

2. U.S. Patent

   Device localization using WiFi/BLE and audio signals

   Amod K. Agrawal, Chaitanya Desai, Lakshmi Venkatraman, and 3 more authors

   Patent Application, United States Patent and Trademark Office, Dec 2023

   U.S. Patent App. 19/248,290, Pending

   Abs Bib

   This patent application discloses a hybrid localization technique using WiFi, Bluetooth Low Energy (BLE), and audio signals. By combining RF-based ranging with acoustic signal processing, the system achieves more accurate and robust indoor localization of smart devices, overcoming limitations of single-modality approaches.

   @patent{agrawal2024wifibleaudio,
     author = {Agrawal, Amod K. and Desai, Chaitanya and Venkatraman, Lakshmi and Bocca, Maurizio and Argyropoulos, Paraskevas and Kim, Wontak},
     title = {Device localization using WiFi/BLE and audio signals},
     month = dec,
     year = {2023},
     location = {United States Patent and Trademark Office},
     number = {18/542,222},
     school = {Patent Application},
     note = {U.S. Patent App. 19/248,290, Pending},
     address = {United States}
   }

2019

1. M.S. Thesis

   HeadTrack: Tracking Head Orientation Using Wireless Signals

   Amod K. Agrawal, and Romit Roy Choudhury

   University of Illinois at Urbana-Champaign, Apr 2019

   Abs DOI Bib Slides

   Estimating and tracking a user’s head orientation is critical for many mobile computing applications. Existing solutions typically rely on infrastructure-based setups, such as cameras or lasers, combined with expensive inertial measurement units (IMUs). These approaches constrain the user to a fixed area and limit both portability and mobility. This work introduces HeadTrack, a necklace-style wearable consisting of a headset and a chest-piece, which estimates head orientation using wireless radio frequency (RF) signals. The core challenge addressed is to accurately measure multiple distances between the chest-mounted unit and the headset using ultra-wideband (UWB) radios. By decoupling head motion from body motion, this design enables both mobility and portability. Although UWB radios offer 1 GHz bandwidth and high-speed clocks, they typically achieve only  10 cm ranging accuracy. HeadTrack improves this to  5 mm by introducing a wired reference path. The transmitted signal is split and sent over both wired and wireless paths, allowing more precise distance estimation. For evaluation, we use ViCon to collect ground truth data. HeadTrack leverages an onboard IMU to resolve phase ambiguity and achieves head orientation tracking with a mean error of 6.5°. The system offers a wearable, occlusion-free, cost-effective solution for head orientation tracking with bounded and non-diverging error.

   @mastersthesis{agrawal2019headtrack,
     title = {HeadTrack: Tracking Head Orientation Using Wireless Signals},
     author = {Agrawal, Amod K. and Choudhury, Romit Roy},
     school = {University of Illinois at Urbana-Champaign},
     month = apr,
     year = {2019},
     doi = {http://hdl.handle.net/2142/105242},
   }

2018

1. Springer EPJ Data Science

   Collective aspects of privacy in the online network

   David Garcia, Mansi Goel, Amod K. Agrawal, and 1 more author

   EPJ Data Science, Jan 2018

   Abs DOI Bib Citations (45)

   Preserving individual control over private information is one of the rising concerns in our digital society. Online social networks exist in application ecosystems that allow them to access data from other services, for example gathering contact lists through mobile phone applications. Such data access might allow social networking sites to create shadow profiles with information about non-users that has been inferred from information shared by the users of the social network. This possibility motivates the shadow profile hypothesis: the data shared by the users of an online service predicts personal information of non-users of the service. We test this hypothesis for the first time on Twitter, constructing a dataset of users that includes profile biographical text, location information, and bidirectional friendship links. We evaluate the predictability of the location of a user by using only information given by friends of the user that joined Twitter before the user did. This way, we audit the historical prediction power of Twitter data for users that had not joined Twitter yet. Our results indicate that information shared by users in Twitter can be predictive of the location of individuals outside Twitter. Furthermore, we observe that the quality of this prediction increases with the tendency of Twitter users to share their mobile phone contacts and is more accurate for individuals with more contacts inside Twitter. We further explore the predictability of biographical information of non-users, finding evidence in line with our results for locations. These findings illustrate that individuals are not in full control of their online privacy and that sharing personal data with a social networking site is a decision that is collectively mediated by the decisions of others.

   @article{garcia2018collective,
     title = {Collective aspects of privacy in the online network},
     author = {Garcia, David and Goel, Mansi and Agrawal, Amod K. and Kumaraguru, Ponnurangam},
     journal = {EPJ Data Science},
     volume = {7},
     number = {1},
     pages = {1--13},
     month = jan,
     year = {2018},
     publisher = {SpringerOpen},
     doi = {10.1140/epjds/s13688-018-0130-3},
   }

2015

1. AAAI HCOMP’15

   On optimizing human-machine task assignments

   Andreas Veit, Michael Wilber, Rajan Vaish, and 46 more authors

   AAAI Conference on Conference on Human Computation and Crowdsourcing, HCOMP ’15, Sep 2015

   Work-in-Progress Posters and Demonstration, Tech Report CR-15-01

   Abs DOI arXiv Bib Citations (4)

   When crowdsourcing systems are used in combination with machine inference systems in the real world, they benefit the most when the machine system is deeply integrated with the crowd workers. However, if researchers wish to integrate the crowd with "off-the-shelf" machine classifiers, this deep integration is not always possible. This work explores two strategies to increase accuracy and decrease cost under this setting. First, we show that reordering tasks presented to the human can create a significant accuracy improvement. Further, we show that greedily choosing parameters to maximize machine accuracy is sub-optimal, and joint optimization of the combined system improves performance.

   @article{veit2015optimizing,
     title = {On optimizing human-machine task assignments},
     author = {Veit, Andreas and Wilber, Michael and Vaish, Rajan and Belongie, Serge and Davis, James and Anand, Vishal and Aviral, Anshu and Chakrabarty, Prithvijit and Chandak, Yash and Chaturvedi, Sidharth and Devaraj, Chinmaya and Dhall, Ankit and Dwivedi, Utkarsh and Gupte, Sanket and Sridhar, Sharath N and Paga, Karthik and Pahuja, Anuj and Raisinghani, Aditya and Sharma, Ayush and Sharma, Shweta and Sinha, Darpana and Thakkar, Nisarg and Vignesh, K Bala and Verma, Utkarsh and Abhishek, Kanniganti and Agrawal, Amod K. and Aishwarya, Arya and Bhattacharjee, Aurgho and Dhanasekar, Sarveshwaran and Gullapalli, Venkata Karthik and Gupta, Shuchita and Jain, Kinjal and Kapur, Simran and Kasula, Meghana and Kumar, Shashi and Kundaliya, Parth and Mathur, Utkarsh and Mishra, Alankrit and Mudgal, Aayush and Nadimpalli, Aditya and Nihit, Munakala Sree and Periwal, Akanksha and Sagar, Ayush and Shah, Ayush and Sharma, Vikas and Sharma, Yashovardhan and Siddiqui, Faizal and Singh, Virender and Yadav, Anurag},
     journal = {AAAI Conference on Conference on Human Computation and Crowdsourcing, HCOMP '15},
     month = sep,
     year = {2015},
     doi = {10.48550/arXiv.1509.07543},
     note = {Work-in-Progress Posters and Demonstration, Tech Report CR-15-01},
   }