About

Joseph Paradiso is the Alexander W. Dreyfoos Professor and the academic head of the Program in Media Arts and Sciences at MIT's Media Lab, where he also directs the Responsive Environments research group. His research focuses on developing sensing technologies that can efficiently capture and process multiple sensing modalities, applying these innovations across diverse domains such as the internet of things, medicine, environmental sensing, space exploration, and artistic expression. Paradiso's work aims to help people better understand the world, express themselves, and connect with others through advanced sensing and interactive technologies. With a background in physics and a PhD in experimental high-energy physics from MIT, Paradiso's early career was marked by pioneering work in wireless wearable sensing. He built systems with embedded sensors capable of real-time data transmission from the human body, including a notable project in 1997 involving shoes embedded with sensors for augmented dance performances. His research has evolved from individual sensing to group applications, such as platforms enabling dance ensembles to create music through collective motion and wearable sensors for sports medicine to assess injury risk and performance. More recently, Paradiso's collaborations with National Geographic Explorers have extended his sensing innovations to study animal behavior and ecosystems, deploying sensors in remote environments to monitor wildlife and environmental conditions. Recognized as an IEEE Fellow for his contributions to wireless wearable sensing and mobile energy harvesting, Paradiso's work exemplifies how foundational research at MIT can lead to impactful applications across art, health, and environmental sciences, emphasizing technology's role in enhancing perception, connectivity, and awareness.

Research topics

• Computer Science
• Artificial Intelligence
• Materials science
• Mechanical engineering
• Engineering
• Immunology
• Virology
• Biochemistry
• Computer vision
• Geology
• Embedded system
• Physics
• Electrical engineering
• Biology
• Acoustics
• Electronic engineering
• Medicine
• Computer hardware

Selected publications

• MindCube V2: Interactive Fidget Device for Emotion Gauging

  2026-04-13

  article
  PublisherDOI

• Are We Doing Enough? Sharing Design and Manufacturing Expertise Across Open Source Hardware Communities

  Bristol Research (University of Bristol) · 2026-02-09

  articleOpen access

  Working with open source hardware designs is often a challenging task. Details about the design and the steps necessary for production may be limited, and bringing the relevant information together often introduces significant educational, institutional and socio-economic barriers. Where explicit support for working with open source hardware design files is provided, its efficacy is difficult to measure. Not only is it often hard for the community to re-create and build on open source designs, it’s also hard for authors of open source projects to know how best to provide suitable help and guidance.<br/><br/>As a research community, could we do more to ensure that information relating to our open source hardware designs is accessible, understandable and effective? Are we using the best processes and channels to empower the widest audience to work with and replicate open source hardware? This workshop will explore and assess methods of information sharing within and between open hardware communities and their efficacy. We will examine industrial, academic and social spheres to construct a multidimensional image of where help can be found, what help is out there, how it can best be leveraged, and what opportunities exist to enhance access to it.

  Publisher

• RGB-NIR Reflectance and 3D Microtopography for Lunar Regolith Analysis Using ToF Imaging Systems

  2026-03-07

  article

  We present an integrated approach for analyzing lunar regolith using co-registered RGB—near-infrared (NIR) reflectance and 3D microtopography derived from a compact Time-of-Flight (ToF) imaging system. The method enables simultaneous surface geometry reconstruction and reflectance-based material characterization under bandwidth, illumination, and payload constraints relevant to lunar surface exploration. Controlled laboratory experiments were conducted on standardized lunar regolith simulants (NU-LHT-2M, LHT-3M, and JSC-1A), systematically varying hydration state and salinity using sodium chloride (NaCl) as a practical analog for volatile residues or permafrost-like deposits. Measurements reveal consistent, monotonic shifts in 850 nm NIR reflectance intensity correlated with increasing moisture content and salt concentration. When combined with visible-band ratios and millimeter-scale surface microtopography, these changes provide repeatable indicators of surface composition, texture, and optical maturity. Radiometric calibration using NIST-traceable photometric targets and linear statistical modeling demonstrates that single-band NIR reflectance, when fused with RGB information and local surface geometry, yields operationally useful proxies for volatile surrogates and compositional contrast. The approach builds on prior demonstrations of low-cost ToF sensing for planetary exploration and is validated in lunar regolith analog environments in support of in-situ prospecting and science operations. These results indicate that compact RGB-NIR ToF imaging systems can serve as low-SWaP, dual-use instruments for lunar science and rover prospecting, providing geometrically co-registered spectral cues that complement multispectral imaging. By enabling material-aware surface assessment directly on 3D reconstructions, the approach supports data fusion workflows and operational decision-making under bandwidth and illumination constraints.

  PublisherDOI

• Are We Doing Enough? Sharing Design and Manufacturing Expertise Across Open Source Hardware Communities

  2026-04-13

  article

  Working with open source hardware designs is often a challenging task. Details about the design and the steps necessary for production may be limited, and bringing the relevant information together often introduces significant educational, institutional and socioeconomic barriers. Where explicit support for working with open source hardware design files is provided, its efficacy is difficult to measure. Not only is it often hard for the community to re-create and build on open source designs, it's also hard for authors of open source projects to know how best to provide suitable help and guidance. As a research community, could we do more to ensure that information relating to our open source hardware designs is accessible, understandable and effective? Are we using the best processes and channels to empower the widest audience to work with and replicate open source hardware? This workshop will explore and assess methods of information sharing within and between open hardware communities and their efficacy. We will examine industrial, academic and social spheres to construct a multidimensional image of where help can be found, what help is out there, how it can best be leveraged, and what opportunities exist to enhance access to it.

  PublisherDOI

• HexSense Lunar Mapping: Deployable 360 Cameras for Panoramic Inspection &amp; 3D Reconstruction

  2026-03-07

  articleSenior author

  The HexSense is a type of low-cost and miniature wireless sensor node that can be ballistically deployed from a rover or lander to the area of interest on the Lunar surface. Upon landing, each HexSense can automatically stand upright to realize a relatively determined orientation, enabling better wireless communication. With a modular design, each HexSense can carry different sensor payloads for different scientific applications. In this paper, we show a novel approach for in-situ panoramic inspection and 3D mapping of the area of interest on the lunar surface using HexSense equipped with custom-designed 360-degree cameras. The unique ballistic deployment mechanism of HexSense allows sensor nodes to be scattered across challenging terrain and automatically orient for optimal data collection. The captured images can be stitched together to panoramic images for in-situ omniview inspection. The images from multiple HexSenses can also be used to reconstruct a 3D model of the area of interest on the lunar surface. This method addresses the challenges of mapping in the hard-to-reach or dangerous areas for a slow-moving rover or a non-movable lander. We described the system design, time synchronization approach, localization approach, panoramic inspection, and 3D reconstruction result, highlighting the potential for scalable, distributed mapping to support future lunar exploration and scientific research.

  PublisherDOI

• A wireless magnetic implant system for continuous neuromuscular sensing

  medRxiv · 2025-11-22

  preprintOpen access

  Abstract Precise measurement of neuromuscular mechanics yields an intuitive control signal for producing synchronous movement with wearable robotics. Practically obtaining such measurements remains challenging as current muscle-sensing technologies excessively compromise between signal fidelity, system complexity, and invasiveness. Here we present a skin-mounted, magnetometer array platform that wirelessly tracks passive 3 mm diameter magnetic beads implanted within human muscle tissue for continuous neuromuscular sensing. The system employs customized high-density sensing electronics with an information-theoretic architecture to achieve sub-millimeter resolution of real-time muscle dynamics at tracking depths of up to 6 cm within the body. We deploy the platform in a first-in-human clinical study to track a constellation of permanently implanted magnets to enable multiple degree-of-freedom neuroprosthetic control. We demonstrate that the wireless muscle state estimation can outperform standard surface and implanted electromyography interfaces to achieve more accurate and responsive neuroprosthetic movement. Further, we successfully supplant electromyography altogether by extending the platform to detect muscle activation through magnetic induction alone. One-Sentence Summary A skin-mounted magnetometer sensing array can wirelessly track 3 mm diameter permanent magnetic implants in humans to provide precise neuromuscular information for improved neuroprosthetic control compared to electromyography.

  PublisherOA PDFDOI

• Purrfect Pitch: Exploring Pitch Interval Learning through an Audio-Haptic Interface

  2025-03-16

  articleOpen access

  AHs 2025, Masdar City, Abu Dhabi, United Arab Emirates

  PublisherOA PDFDOI

• Creating Immersive Digital Twins of Terrestrial Planetary Analogs With Multimodal Sensing and Game Engines for Virtual Exploration

  IEEE Pervasive Computing · 2025-10-01

  articleSenior author

  Creating cross-reality applications of planetary analog environments supports scientific exploration and mission planning by offering a safe and cost-effective way to explore remote terrains. We present a pipeline that integrates physical and virtual data through 3D reconstruction, environmental sensing, and interactive real-time rendering in game engines. The approach was validated at two analog sites in Svalbard, Norway, and Lanzarote, Spain, using UAV photogrammetry, smartphone LiDAR, RGB imagery, and environmental and seismic sensors. In Svalbard, we reconstructed water-indicating terrain in Unity3D. In Lanzarote, we visualized a lava tube with integrated seismic and atmospheric data in Unreal Engine. The environments are explorable in both desktop and VR modes. By combining consumer hardware with multimodal sensing, we demonstrate a flexible method for generating immersive digital twins. We discuss low-cost tools for analog fieldwork, outline design considerations for integration and visualization, and provide recommendations for future cross-reality deployments in science and exploration contexts.

  PublisherDOI

• 45 SPS Members Elevated to Fellow [Society News]

  IEEE Signal Processing Magazine · 2025-11-01

  article
  PublisherDOI

• Internet of Things (IoT): Studying the Integration of Everyday Objects with the Internet and the Implications for Human Lives

  Springer series in adaptive environments · 2025-01-01

  book-chapter
  PublisherDOI

Frequent coauthors

• U. Becker

  European Organization for Nuclear Research

  60 shared

• F. Vannucci

  Sorbonne Université

  57 shared

• H. B. Newman

  Institute of High Energy Physics

  53 shared

• J.D. Burger

  Massachusetts Institute of Technology

  52 shared

• D. Luckey

  ETH Zurich

  51 shared

• James G. Branson

  University of California System

  50 shared

• G. Herten

  University of Freiburg

  45 shared

• M. J. White

  45 shared

Labs

• Apeiron LabsPI

Awards & honors

• IEEE Fellow (2026)