About

Ashutosh Dhekne is an associate professor in the School of Computer Science at Georgia Tech. His research focuses on ultra-wideband (UWB) localization and sensing, the Internet of Things, and mobile computing—bridging algorithms, systems, and hardware prototypes. He received the NSF CAREER Award in 2022 for advances in wireless localization and sensing. Ashutosh earned his Ph.D. from the University of Illinois at Urbana–Champaign in 2019. He is a BBISS Fellow.

Research topics

• Computer Science
• Artificial Intelligence
• Computer Security
• Telecommunications
• Engineering
• Electrical engineering
• Computer vision
• Real-time computing
• Computer graphics (images)
• Computer network
• Embedded system

Selected publications

• Flash-Fusion: Enabling Expressive, Low-Latency Queries on IoT Sensor Streams with LLMs

  ArXiv.org · 2025-11-14

  preprintOpen access

  Smart cities and pervasive IoT deployments have generated interest in IoT data analysis across transportation and urban planning. At the same time, Large Language Models offer a new interface for exploring IoT data - particularly through natural language. Users today face two key challenges when working with IoT data using LLMs: (1) data collection infrastructure is expensive, producing terabytes of low-level sensor readings that are too granular for direct use, and (2) data analysis is slow, requiring iterative effort and technical expertise. Directly feeding all IoT telemetry to LLMs is impractical due to finite context windows, prohibitive token costs at scale, and non-interactive latencies. What is missing is a system that first parses a user's query to identify the analytical task, then selects the relevant data slices, and finally chooses the right representation before invoking an LLM. We present Flash-Fusion, an end-to-end edge-cloud system that reduces the IoT data collection and analysis burden on users. Two principles guide its design: (1) edge-based statistical summarization (achieving 73.5% data reduction) to address data volume, and (2) cloud-based query planning that clusters behavioral data and assembles context-rich prompts to address data interpretation. We deploy Flash-Fusion on a university bus fleet and evaluate it against a baseline that feeds raw data to a state-of-the-art LLM. Flash-Fusion achieves a 95% latency reduction and 98% decrease in token usage and cost while maintaining high-quality responses. It enables personas across disciplines - safety officers, urban planners, fleet managers, and data scientists - to efficiently iterate over IoT data without the burden of manual query authoring or preprocessing.

  PublisherOA PDFDOI

• Ultra-wideband (UWB) Radios for Spatial Intelligence

  GetMobile Mobile Computing and Communications · 2025-01-20

  article1st authorCorresponding

  Length is one of the seven fundamental physical quantities. For several centuries we have measured distances using calibrated physical objects, and more recently using light, sound, and radio waves. These measurements and the tools we use have enabled advances in several different domains, from the construction industry to space travel, from GPS localization to tracking of airplanes. With advances in electronics, clocks, miniaturization, and development of new algorithms, wireless distance measurement has now become possible. Measuring distances using wireless sensors offers the option of locating objects across rooms, through walls, and without visual line-of-sight. The measurement accuracy improves with larger bandwidth, which has resulted in the ultra-wideband (UWB) radio technology gaining significant traction. Seeing an opportunity in this capability, smartphone manufacturers such as Google and Apple have incorporated UWB radios in their offerings, setting the stage for future innovation using this versatile technology. [10] In this article we look beyond the UWB object-finding use-case and explore how a modest radio can transform mobile computing for decades to come.

  PublisherDOI

• uThaw: Ultra-Wideband Wireless Solid–Liquid State Transition Sensor to Detect Thawing of Food

  IEEE Sensors Letters · 2025-01-31 · 2 citations

  articleSenior author

  This letter explores the potential use of ultra-wideband (UWB) wireless sensors in detecting the transition of food items from a frozen state to thawed state (or vice versa), marking a significant advancement in monitoring the thawing process. By exploiting the drastic change in complex permittivity at microwave frequencies (we use frequencies close to 4 GHz) during the solid–liquid state transition, this letter introduces a novel approach in ensuring frozen food safety and in enhancing the efficiency of the frozen food industry and cold chain transportation. The developed system, capable of operating through food packaging, offers a noninvasive, cost-effective solution for real-time monitoring, addressing the limitations of conventional temperature-based and timing-based methods widely used today in household and professional cooking and in the food industry. Our findings from the raw UWB channel impulse responses and computed similarity scores indeed show significant promise and validate the feasibility of the proposed system with various real-world applications.

  PublisherDOI

• Bringing Context to the Underserved: Rethinking Context-Aware Design to Bridge the Digital Divide

  2025-07-18 · 2 citations

  article
  PublisherDOI

• UWBKey: Using Contrastive Learning for Efficient Secure Key Generation in UWB

  Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · 2025-12-02 · 1 citations

  articleOpen accessSenior author

  Ultra-wideband radios are being used for a multitude of localization and ranging applications, from object finding using Apple AirTags, to industrial precise asset localization, to keyless car doors. UWB's ubiquitous presence is fueled, in part, by its effectiveness in separating the wireless signal's first path from multipath and its co-existence with Wi-Fi and other wireless technologies. Yet, today UWB-based devices are still required to be “paired” with a controller device which can gain exclusive access to communicate with that device. This pairing requirement restricts truly ubiquitous and pervasive use-cases for UWB. One of the primary reasons for the pairing requirement is to enable secure UWB communication, including the communication about the ranging parameters and timestamps which then enables secure UWB ranging. Today, this pairing is performed using an out-of-band communication mechanism based on Bluetooth, or Wi-Fi. In this work, we show that an out of band communication is not necessary for establishing secure keys for UWB communication. In fact, the channel impulse response obtained by every UWB device when receiving a packet can itself be leveraged to generate a symmetric secure key. We call our system, UWBKey , and have fully implemented it on a commercial off-the-shelf UWB chip (Qorvo DWM1000). We show that the channel impulse response (CIR) is reciprocal for a pair of communicating devices, and that we can derive a key from the CIR that is close to random. In contrast, an eavesdropper even when nearby cannot obtain the same CIR and therefore cannot infer the key.; AB@This paper develops an important component of the whole UWB ecosystem within the constraints of today's UWB hardware (in terms of available bandwidth), practical environments (in terms of representative public places we could access), and practical device density (with assumptions about how near a malicious device could get). By no means do we believe we have solved the secure communication problem fully, but we are convinced that our open-data, open-hardware, open-software approach in this paper will enable a new thought process among researchers enabling UWB to stand on its own as a secure radio communication and ranging technology, without depending on out-of-band solutions.

  PublisherDOI

• ObjectTrack: 6DoF Object Tracking Through UWB-IMU Fusion

  2025-09-15

  articleSenior author

  This paper presents a UWB-IMU fusion approach to obtain location and orientation of an object in 6 degrees of freedom at the room level, without use of optical motion capture systems. When tested with different human movement patterns such as walking, running, jumping, and swirling on a wheeled chair, we obtain less than 10cm of 3D localization error and under 5° of orientation error at the 90<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> percentile. We expect our system, called ObjectTrack, to enable spatial audio and interaction for VR/AR applications, enable precision tracking of objects, and for localization of robotic motion systems. ObjectTrack significantly reduces the cost barrier by about 50× compared to popular motion capture systems.

  PublisherDOI

• Position Paper: MoViz: UWB-IMU Sensors as Input to Body-Movement Visualizations

  2024-06-03

  articleOpen accessSenior author

  Generating art through body-movements has widespread utility spanning several fields including physical rehabilitation post-surgery, to entertainment, to health and well-being. This paper introduces a new modality available for researchers via distance measurements using ultra-wideband (UWB) devices, in addition to the traditional IMU based approaches. As a versatile tool, UWB allows rich information to be garnered while consuming minimal data bandwidth and requiring minimal computation that can be easily performed on a microcontroller. We showcase some of the art we have generated using this platform and call upon the research community to explore the topic further.

  PublisherOA PDFDOI

• Poster: Envisioning a UWB-based Local Human-Machine Interface

  2024-12-04

  articleOpen accessSenior author

  In this work, we present a secure Internet-of-Things framework enabled by UWB, which allows users to wirelessly interact with any public and home appliances in a more secure, seamless and efficient manner. we envision an ecosystem of appliances and mobile devices equipped with UWB transceivers, where the appliances are capable of imposing location-based access control in precise and configurable zones and ensuring security against eavesdropping and hijacking by malicious attackers. We propose a location-based access control algorithm that is robust against range spoofing. Further, we propose the network and application layer protocols that enable one generic user-end application to access the control interface of any appliances and adapt to context.

  PublisherOA PDFDOI

• Thaw: A UWB-based Ice-Water State Detector

  2024-02-20 · 1 citations

  articleSenior author

  This work explores the use of wireless signals to distinguish between solid and liquid states of water inside enclosed spaces such as freezers and microwave ovens. This proof-of-concept system, utilizes ultra-wideband (UWB) technology, demonstrating how changes in wireless reflection patterns can indicate transition between ice and water states.

  PublisherDOI

• uThaw: Ultra-wideband Wireless Solid-Liquid State Transition Sensor to Detect Thawing of Food

  2024-05-09

  preprintOpen accessSenior author

  This paper explores the potential use of ultra-wideband (UWB) wireless sensors in detecting the transition of food items from a frozen state to thawed state (or vice-versa), marking a significant advancement in monitoring the thawing process. By exploiting the drastic change in complex permittivity at microwave frequencies (we use frequencies close to 4GHz) during the solid-liquid state transition, this paper introduces a novel approach in ensuring frozen food safety, and in enhancing the efficiency of the frozen food industry and cold chain transportation. The developed system, capable of operating through food packaging, offers a non-invasive, cost-effective solution for real-time monitoring, addressing the limitations of conventional temperature-based and timing-based methods widely used today in household and professional cooking and in the food industry. Our findings from the raw UWB channel impulse responses (CIR) and computed similarity scores indeed show significant promise and validate the feasibility of the proposed system with various realworld applications.

  PublisherOA PDFDOI

Recent grants

• RAPID: 6Fit-a-Part: A Device for Physical Distancing

  NSF · $100k · 2020–2021

• CAREER: Closing the Gaps in UWB Localization and Sensing; Algorithms, Architectures, and Prototypes

  NSF · $650k · 2022–2027

Frequent coauthors

• Romit Roy Choudhury

  13 shared

• Haige Chen

  Renji Hospital

  9 shared

• Yifeng Cao

  Georgia Institute of Technology

  9 shared

• Mahanth Gowda

  Pennsylvania State University

  9 shared

• Mostafa Ammar

  Georgia Institute of Technology

  7 shared

• Umakishore Ramachandran

  Georgia Institute of Technology

  5 shared

• Manasvini Sethuraman

  Georgia Institute of Technology

  4 shared

• Anirudh Sarma

  Georgia Institute of Technology

  4 shared

Education

• PhD, Computer Science

  University of Illinois at Urbana-Champaign

  2019

• MTech, Computer Science

  Indian Institute of Technology Bombay

  2009

• Bachelor of Engineering, Computer Science

  Savitribai Phule Pune University

  2006

Awards & honors

• NSF CAREER Award (2022)
• BBISS Fellow