About

Amin Arbabian is a researcher leading the Arbabian Lab at Stanford University, where the focus is on designing and building end-to-end intelligent sensing systems that bridge the physical and digital worlds. His work spans the entire stack, from physics and high-performance/high-frequency circuits to algorithms and edge intelligence, integrating RF, ultrasound, and optical modalities. His research aims to create new capabilities for autonomy, healthcare, the Internet of Things, and scientific discovery by tightly coupling hardware, signal processing, and inference, leveraging multi-physics and multi-disciplinary approaches that combine electromagnetics, acoustics, and hybrid wave interactions with custom electronic design and machine learning to develop physical AI sensing systems.

Research topics

• Computer Science
• Physics
• Artificial Intelligence
• Telecommunications
• Remote sensing
• Electronic engineering
• Oceanography
• Optics
• Engineering
• Geology
• Optoelectronics
• Acoustics

Selected publications

• Wireless soft implantable pressure sensors based on liquid metal with large-area mapping and adaptive implantation capabilities

  Proceedings of the National Academy of Sciences · 2026-02-17 · 1 citations

  articleOpen access

  Innovations in soft materials have advanced the development of implantable devices for pressure monitoring, but fabrication and integration challenges remain, such as limited patterning resolution and poor scalability, hindering their miniaturization and wireless sensing capabilities. This study introduces methods and advantageous features of incorporating liquid metal into microfabriated, wireless soft implantable pressure sensors and wearable readout systems for large-area pressure mapping and adaptive implantation with autonomous folding and self-healing capabilities. Eutectic gallium-indium, a type of liquid metal, serves as both the deformable electrode for capacitive sensors and a low-resistance conductor for inductors. It is integrated into a thin-film, battery-free, inductive-capacitive wireless sensing platform. Scalable wireless sensor arrays are created through microfabrication for large-area pressure mapping. The wireless pressure sensor is also integrated with soft ferromagnetic and self-healing layers in cuff-type sensors to allow for autonomous folding in response to external magnets, eliminating the need for suturing. In addition, a miniaturized wearable readout system integrated into medical gloves enables wireless and real-time pressure monitoring. The presented wireless soft pressure-sensing method with large-area mapping and secure implantation capabilities offers alternatives to conventional medical tools for intraoperative monitoring and examinations.

  PublisherDOI

• The 2025 RFIC Symposium

  IEEE Microwave Magazine · 2025-04-11

  articleSenior author

  Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.

  PublisherDOI

• Underwater Imaging with an Air-Coupled Piezoelectric Micromachined Ultrasonic Transducer Phased Array

  2025-09-15

  articleSenior author

  Oceans cover over 70% of the Earth’s surface and are very important for various ecological processes, but have been largely unmapped. Thus, it is imperative to fill this gap by providing high-resolution, high-coverage underwater imaging. Airborne sonar is one such emerging approach for underwater imaging that measures acoustic signals propagating from water to air. Due to significant attenuation at the air-water interface, highly sensitive sensors such as piezoelectric micromachined ultrasonic transducers (PMUTs) are necessary, as conventional ultrasonic sensors do not provide adequate sensitivity. Recent developments in PMUT technology have led to the availability of commercial-off-the-shelf (COTS) sensors that are both cost-effective and highly sensitive, facilitating efficient hardware scaling and seamless integration. Consequently, we leverage these advantages to develop a 1D linear PMUT phased array for scalable airborne sonar imaging, with experimental validation through array characterization and underwater target imaging.

  PublisherDOI

• A 40.68-MHz Fully-Integrated Voltage/Current-Mode Dual-Output PMU for Wireless Neural Implants

  IEEE Transactions on Biomedical Circuits and Systems · 2025-07-22

  article

  This paper presents a fully-integrated single-input dual-output power management unit operating both in voltage/current modes for powering mm-scale wireless neural implants. The chip operates in voltage mode most of the time, using an active full-wave rectifier to regulate a low-voltage, high-load output with high power efficiency and low output ripple ($<$32 $\mathrm{mV}_{\mathrm{pp}}$). It switches to current mode rectification when generating a high-voltage, low-load output. This dual-mode operation allows for flexible power distribution and configurable voltage ratios between the two outputs. The selected 40.68 MHz operating frequency reduces the required capacitances for input impedance matching and output filtering, enabling on-chip integration; the only external component is the receiver coil. A novel resonance breakup switch compatible with full-wave rectification ensures a smooth cold start-up of the chip without any external voltage supply. The chip was fabricated using 40-nm CMOS technology with an active area of 1.18 $\mathrm{mm^{2}}$ and was tested in a wireless power link. Measurement results demonstrate that the chip can simultaneously regulate two outputs, $V_{LV}$ = 1 V and $V_{HV}$ = 2 V, with a tested maximum output power of 10 mW and 32.6 µW on $V_{LV}$ and $V_{HV}$, respectively. At the optimal output power condition ($P_{LV}$ = 4.4$\sim$6.7 mW), the system achieves a peak power conversion efficiency of 85.87% and a peak end-to-end efficiency of 17.32% when regulating $V_{LV}$. The end-to-end efficiency drops by only 2.38% when regulating both outputs with $R_{LV}$ = 225 $\Omega$ and $R_{HV}$ = 400 k$\Omega$.

  PublisherDOI

• High-Resolution Gait Micro-Doppler Synthesis from Videos Over Diverse Trajectories

  2025-03-12 · 1 citations

  articleSenior author

  In recent years, there has been increasing interest in human body motion analysis, with applications in activity classification, human gait analysis, and human intent recognition. Among the various methods, millimeter-wave (mmWave)-based approaches have become popular due to their inherent contactless and privacy-preserving properties, as well as their resilience to lighting, weather conditions, and measurement distance. However, mmWave-based human motion analysis is still in its early stages, primarily due to the limited availability of large-scale datasets. This is particularly true for tasks that require high-resolution motion measurements. In this work, we introduce a novel method for synthesizing high-resolution mmWave datasets directly from videos. The proposed approach is well-suited for applications requiring very high-resolution Doppler signature simulations, such as analyzing subtle hand motions of walking pedestrians. We achieve this by employing an adversarial training strategy combined with custom task-specific loss functions that enhance the micro-motion signatures in the hands and legs of walking pedestrians. This work is the first to design and validate a high-resolution synthesized Doppler dataset of walking activities across multiple trajectories and subjects.

  PublisherDOI

• MVDoppler-Pose: Multi-Modal Multi-View mmWave Sensing for Long-Distance Self-Occluded Human Walking Pose Estimation

  2025-06-10 · 4 citations

  articleSenior author

  One of the main challenges in reliable camera-based 3D pose estimation for walking subjects is to deal with self-occlusions, especially in the case of using low-resolution cameras or at longer distance scenarios. In recent years, millimeter-wave (mmWave) radar has emerged as a promising alternative, offering inherent resilience to the effect of occlusions and distance variations. However, mmWave-based human walking pose estimation (HWPE) is still in the nascent development stages, primarily due to its unique set of practical challenges including the quality of the observed radar signal dependent on the subject’s motion direction. This paper introduces the first comprehensive study comparing mmWave radar to camera systems for HWPE, highlighting its utility for distance-agnostic and occlusion-resilient pose estimation. Building upon mmWave’s unique advantages, we address its intrinsic directionality issue through a new approach—the synergetic integration of multi-modal, multi-view mmWave signals, achieving robust HWPE against variations both in distance and walking direction. Extensive experiments on a newly curated dataset not only demonstrate the superior potential of mmWave technology over traditional camera-based HWPE systems, but also validate the effectiveness of our approach in over-coming the core limitations of mmWave HWPE.

  PublisherDOI

• Asynchronous sensor fusion for multiple object tag-less activity tracking in manufacturing

  2024-06-07

  articleSenior author

  Centralized fusion ensures minimal information loss and maximizes the effectiveness of state estimation. Statistically, it is the optimal solution for all sensor fusion configurations. In this paper, we introduce a local-sensor-driven asynchronous low-level centralized fusion methodology that seamlessly integrates radar and camera data at the level of detections from each sensor. For a local-sensor-driven asynchronous system, detections from the two sensing modalities with different sampling rates are transmitted to a centralized filter, which is updated whenever it receives a measurement. We implemented the proposed algorithm and validated the results using real data from manufacturing and industrial work sites. The data was obtained by Plato System’s Argus perception system, which combines high-resolution imaging mm-wave radar with camera sensors to provide indoor and outdoor activity tracking. We further compare the fusion results with vision-only MOT, as well as track-level fusion (track-to-track fusion).

  PublisherDOI

• 6.1 12Mb/s 4×4 Ultrasound MIMO Relay with Wireless Power and Communication for Neural Interfaces

  2024-02-18 · 8 citations

  articleSenior author

  Neural interfaces, including retinal and brain implants, require increasing post-compression uplink data rates $\gt10$ Mb/s to send data from larger and denser recording arrays. In retinal implants, electrical recording data from large arrays can be leveraged to identify cell types, stimulation thresholds, and locations to construct better stimulation patterns. Among current uplink solutions, RF has the highest data rate but has limited depth inside the body for small transmitter (Tx) sizes [1, 2]. Galvanic solutions can achieve data rates $\sim10$ Mb/s [3] with low power and small form factor but have limited scaling in number of implants and data rate due to their omnidirectionality. Ultrasound (US) in implant applications can operate at depth with small transducers but has so far been limited to data rates <1 Mb/s [4, 5] due to power and area constraints.

  PublisherDOI

• Birefringence-free photoelastic modulator with centimeter-square aperture operating at 2.7 MHz with sub-watt drive power

  Optics Letters · 2024-08-13 · 4 citations

  articleSenior author

  Photoelastic modulators are optical devices with a broad range of applications. These devices typically utilize a transverse interaction mechanism between acoustic and optical waves, resulting in a fundamental trade-off between the input aperture and the modulation frequency. Commercially available modulators with centimeter-square apertures have operating frequencies in the vicinity of 50 kHz. In this work, we experimentally demonstrate a birefringence-free photoelastic modulator operating at approximately 2.7 MHz with a centimeter-square aperture, increasing the operating frequency substantially compared to existing approaches. Using the modulator and polarizers, we demonstrate close to π radians polarization modulation amplitude with sub-watt drive power, translating to nearly 100% intensity modulation efficiency at the fundamental (2.7 MHz) and second-harmonic (5.4 MHz) frequencies.

  PublisherDOI

• Productivity Enhancement in Semiconductor Manufacturing with Ai-Enabled Operations Digital Twin Platform

  2024-10-22 · 2 citations

  article1st authorCorresponding

  There is a significant need to optimize human-machine collaboration in electronics manufacturing, where high labor and equipment costs make downtimes costly. Although digital twins offer promising opportunities, only <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$26 \%$</tex> of companies have made substantial progress in their initiatives. This paper presents an AI-enabled operations digital twin platform that digitizes operator activities and integrates them with machine and other factory data to create a comprehensive digital replica of manufacturing processes. Utilizing radar, computer vision, and sensor fusion on the edge, the system enables precise indoor tracking without the need for wearable tags. The combination of digitized activity and factory operations data is then analyzed using AI to generate actionable insights about adverse operational events such as downtimes and their root-causes Deployed in several semiconductor and electronics manufacturing companies, the platform has been used to identify opportunities to significantly improve production efficiency, reduce downtimes, and minimize operational variability. These results underscore the importance of incorporating operator data for optimal manufacturing operations.

  PublisherDOI

Recent grants

• In vivo Wireless Sensors for Gut Redox Monitoring to Understand Host and Microbe Physiology

  NIH · $434k · 2021–2023

• NeTS: Large: Collaborative Research: GigaNets: A Path to Experimental Research in Millimeter Wave Networking

  NSF · $480k · 2015–2020

• CAREER: Ultrasonically-Powered Smart Medical Implants for Monitoring and Modulating Local Physiology

  NSF · $500k · 2015–2021

• A Wireless, Implantable Microdevice for Closed-Loop Drug Delivery to Prevent the Morbidity of Diabetes Therapy-Induced Hypoglycemia

  NIH · $2.2M · 2018–2022

Frequent coauthors

• Jayant Charthad

  28 shared

• Marcus J. Weber

  Samford University

  25 shared

• Ting Chia Chang

  Stanford University

  23 shared

• Hao Nan

  Tianjin University of Traditional Chinese Medicine

  22 shared

• Jun‐Chau Chien

  21 shared

• Ali M. Niknejad

  19 shared

• Max L. Wang

  Stanford University

  16 shared

• Spyridon Baltsavias

  Stanford University

  16 shared

Labs

• Arbabian LabPI

  Circuit/system design in mm-Wave and THz, Biomedical, and Ultra-Low Power Electronic sensors

Education

• Ph.D., Electrical Engineering

  Stanford University

  2007

• M.S., Electrical Engineering

  Stanford University

  2003

• B.S., Electrical Engineering

  University of California, Los Angeles

  2001