Minah Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Minah Lee has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
Saibal Mukhopadhyay
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
4 shared publications- Saibal Mukhopadhyay
Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
3 shared publications - Burhan Ahmad Mudassar
Georgia Institute of Technology
2 shared publications - Edward Lee
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
2 shared publications - Hakki Mert Torun
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
2 shared publications Madhavan Swaminathan
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332 USA
2 shared publications
Latest Publications
- 1 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 9 (11) , 1-4
Lightweight Local Feature Restoration for Reliable Energy-Efficient Feature Extraction
- 1 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 9 (11) , 1-4
Lightweight Local Feature Restoration for Reliable Energy-Efficient Feature Extraction
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Artificial Intelligence
- Vol. 2 (2) , 139-149
Energy-Efficient and Reliable Sensor Platform Based on Analog-to-Feature Extraction
This paper introduces the Artificial Intelligence-Driven Energy-efficient Analog-to-Learning system (AIDEAL), an AI-enabled in-sensor computing platform designed to enhance energy efficiency in edge devices through in situ data processing. AIDEAL extracts task-relevant features from analog inputs, significantly reducing the data volume for digitization and transmission, by utilizing analog compute-in-memory (ACIM). AIDEAL demonstrates substantial benefits across various applications including image reconstruction, image classification, and object detection. Notably, it achieves 60.2%-80.2% reduction in sensor energy for image classification compared to traditional methods of digitizing and transmitting full input data. We also examine AIDEAL under different feature reduction rates and variations in voltage, and temperature in ACIM. Furthermore, we introduce two algorithmic strategies, feature restoration (FR) and hamming weight based energy aware quantization (HEQA), which enhance accuracy and energy efficiency. These methods result in less than 2% accuracy drop and 37% energy savings compared to baseline, respectively.
- 4 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Artificial Intelligence
- Vol. 2 (2) , 139-149
Energy-Efficient and Reliable Sensor Platform Based on Analog-to-Feature Extraction
This paper introduces the Artificial Intelligence-Driven Energy-efficient Analog-to-Learning system (AIDEAL), an AI-enabled in-sensor computing platform designed to enhance energy efficiency in edge devices through in situ data processing. AIDEAL extracts task-relevant features from analog inputs, significantly reducing the data volume for digitization and transmission, by utilizing analog compute-in-memory (ACIM). AIDEAL demonstrates substantial benefits across various applications including image reconstruction, image classification, and object detection. Notably, it achieves 60.2%-80.2% reduction in sensor energy for image classification compared to traditional methods of digitizing and transmitting full input data. We also examine AIDEAL under different feature reduction rates and variations in voltage, and temperature in ACIM. Furthermore, we introduce two algorithmic strategies, feature restoration (FR) and hamming weight based energy aware quantization (HEQA), which enhance accuracy and energy efficiency. These methods result in less than 2% accuracy drop and 37% energy savings compared to baseline, respectively.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (12) , 11317-11327
Energy Efficient Pixel-Parallel Read-Out Circuits for Digital Image Sensors Using Cross-Layer Pixel Depth Control
Digital pixel sensors with pixel-parallel readout provide opportunities for high frame rate and high resolution in 3D integrated camera platforms with in-sensor machine learning. The active energy consumption of the digital pixels at the sensor front end is a key performance parameter for applications involving smart cameras. This work proposes a novel digital pixel design that utilizes cross-layer feedback to improve active energy efficiency of pixels in image sensors. An algorithm-circuit co-simulation of $336\times256$ image sensor in 65nm CMOS for an object detection task shows 58% active energy saving in digital pixel circuits, while maintaining accuracy of end task.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (12) , 11317-11327
Energy Efficient Pixel-Parallel Read-Out Circuits for Digital Image Sensors Using Cross-Layer Pixel Depth Control
Digital pixel sensors with pixel-parallel readout provide opportunities for high frame rate and high resolution in 3D integrated camera platforms with in-sensor machine learning. The active energy consumption of the digital pixels at the sensor front end is a key performance parameter for applications involving smart cameras. This work proposes a novel digital pixel design that utilizes cross-layer feedback to improve active energy efficiency of pixels in image sensors. An algorithm-circuit co-simulation of $336\times256$ image sensor in 65nm CMOS for an object detection task shows 58% active energy saving in digital pixel circuits, while maintaining accuracy of end task.
- 5 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (12) , 13794-13804
Adaptive Camera Platform Using Deep Learning-Based Early Warning of Task Failures
There is a growing interest in deploying complex deep neural networks (DNN) in smart sensors to extract task-specific information from real-time sensor data. This paper presents an adaptive sensor that uses a novel light-weight deep learning platform, WarningNet, to estimate potential task failures due to perturbations in the sensor data and control its read-out-circuit (ROIC) parameters such as sampling resolution and operating voltage. Simulation results show that WarningNet can provide early warning of the performance degradation of various tasks within a fraction of the time required for the task to complete. Moreover, the early warning guided adaptive control of sampling resolution and operating voltage of the ROIC of a digital pixel sensor shows potential of dynamic trade off between task accuracy and sensor energy/bandwidth.
- 5 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 21 (12) , 13794-13804
Adaptive Camera Platform Using Deep Learning-Based Early Warning of Task Failures
There is a growing interest in deploying complex deep neural networks (DNN) in smart sensors to extract task-specific information from real-time sensor data. This paper presents an adaptive sensor that uses a novel light-weight deep learning platform, WarningNet, to estimate potential task failures due to perturbations in the sensor data and control its read-out-circuit (ROIC) parameters such as sampling resolution and operating voltage. Simulation results show that WarningNet can provide early warning of the performance degradation of various tasks within a fraction of the time required for the task to complete. Moreover, the early warning guided adaptive control of sampling resolution and operating voltage of the ROIC of a digital pixel sensor shows potential of dynamic trade off between task accuracy and sensor energy/bandwidth.
- 19 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 10 (4) , 444-457
Cross-Layer Noise Analysis in Smart Digital Pixel Sensors With Integrated Deep Neural Network
Digital pixel based image sensors with embedded deep neural network (DNN) allow many mission critical surveillance applications. However, image noise caused by variations and non-idealities in the sensor aggravates the quality of image and further degrades the performance of a DNN. We propose a digital pixel-DNN cross-layer simulation methodology for accurate training and evaluation of a DNN under image noise induced from sensors. In particular, this paper focuses on the image noise derived from device mismatches in digital pixel circuits with 3D integrated and pixel-parallel readout circuits, and studies the effect of the resulting image noise on the accuracy of a DNN. The simulation results show that the device mismatch in the digital pixel creates distinct noise structure on output image and should be accurately considered while training a DNN. We also present design space explorations using our cross-layer simulation.
- 19 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 10 (4) , 444-457
Cross-Layer Noise Analysis in Smart Digital Pixel Sensors With Integrated Deep Neural Network
Digital pixel based image sensors with embedded deep neural network (DNN) allow many mission critical surveillance applications. However, image noise caused by variations and non-idealities in the sensor aggravates the quality of image and further degrades the performance of a DNN. We propose a digital pixel-DNN cross-layer simulation methodology for accurate training and evaluation of a DNN under image noise induced from sensors. In particular, this paper focuses on the image noise derived from device mismatches in digital pixel circuits with 3D integrated and pixel-parallel readout circuits, and studies the effect of the resulting image noise on the accuracy of a DNN. The simulation results show that the device mismatch in the digital pixel creates distinct noise structure on output image and should be accurately considered while training a DNN. We also present design space explorations using our cross-layer simulation.
- Top Cited Papers
- 24 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 28 (11) , 2424-2437
Architecture, Chip, and Package Codesign Flow for Interposer-Based 2.5-D Chiplet Integration Enabling Heterogeneous IP Reuse
A new trend in system-on-chip (SoC) design is chiplet-based IP reuse using 2.5-D integration. Complete electronic systems can be created through the integration of chiplets on an interposer, rather than through a monolithic flow. This approach expands access to a large catalog of off-the-shelf intellectual properties (IPs), allows reuse of them, and enables heterogeneous integration of blocks in different technologies. In this article, we present a highly integrated design flow that encompasses architecture, circuit, and package to build and simulate heterogeneous 2.5-D designs. Our target design is 64-core architecture based on Reduced Instruction Set Computer (RISC)-V processor. We first chipletize each IP by adding logical protocol translators and physical interface modules. We convert a given register transfer level (RTL) for 64-core processor into chiplets, which are enhanced with our centralized network-on-chip. Next, we use our tool to obtain physical layouts, which is subsequently used to synthesize chip-to-chip I/O drivers and these chiplets are placed/routed on a silicon interposer. Our package models are used to calculate power, performance, and area (PPA) and reliability of 2.5-D design. Our design space exploration (DSE) study shows that 2.5-D integration incurs 1.29x power and 2.19x area overheads compared with 2-D counterpart. Moreover, we perform DSE studies for power delivery scheme and interposer technology to investigate the tradeoffs in 2.5-D integrated chip (IC) designs.
- 24 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 28 (11) , 2424-2437
Top Cited PapersArchitecture, Chip, and Package Codesign Flow for Interposer-Based 2.5-D Chiplet Integration Enabling Heterogeneous IP Reuse
A new trend in system-on-chip (SoC) design is chiplet-based IP reuse using 2.5-D integration. Complete electronic systems can be created through the integration of chiplets on an interposer, rather than through a monolithic flow. This approach expands access to a large catalog of off-the-shelf intellectual properties (IPs), allows reuse of them, and enables heterogeneous integration of blocks in different technologies. In this article, we present a highly integrated design flow that encompasses architecture, circuit, and package to build and simulate heterogeneous 2.5-D designs. Our target design is 64-core architecture based on Reduced Instruction Set Computer (RISC)-V processor. We first chipletize each IP by adding logical protocol translators and physical interface modules. We convert a given register transfer level (RTL) for 64-core processor into chiplets, which are enhanced with our centralized network-on-chip. Next, we use our tool to obtain physical layouts, which is subsequently used to synthesize chip-to-chip I/O drivers and these chiplets are placed/routed on a silicon interposer. Our package models are used to calculate power, performance, and area (PPA) and reliability of 2.5-D design. Our design space exploration (DSE) study shows that 2.5-D integration incurs 1.29x power and 2.19x area overheads compared with 2-D counterpart. Moreover, we perform DSE studies for power delivery scheme and interposer technology to investigate the tradeoffs in 2.5-D integrated chip (IC) designs.
- 14 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 10 (1) , 111-122
Automated I/O Library Generation for Interposer-Based System-in-Package Integration of Multiple Heterogeneous Dies
System-in-Package (SiP) integration of multiple dies in a single package can achieve much higher performance than on-board integration of ICs while reducing the design cost/effort compared to large System-on-Chips (SoCs). However, a major challenge in design of SiPs with many dies is automated design and insertion of Input/Output (I/O) cells to minimize energy and delay of the wire traces. This paper presents an automated cell library generation flow for all-digital I/O circuits for SiP integration. Given parameterized models of SiP wire traces, our method automatically designs, optimizes, and generates layouts of I/O cells for delay/energy minimization. The proposed flow is demonstrated on interposer based SiP integration considering 28nm CMOS technology and 65nm BEOL technology. Given a multi-die SiP design and associated interposer wire-traces, this paper demonstrates that automated I/O library cell generation can reduce maximum die-to-die communication delay or energy. We demonstrate the proposed flow for various interposer parameters and SiP designs to show the feasibility of chip-interposer co-design.
- 14 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 10 (1) , 111-122
Automated I/O Library Generation for Interposer-Based System-in-Package Integration of Multiple Heterogeneous Dies
System-in-Package (SiP) integration of multiple dies in a single package can achieve much higher performance than on-board integration of ICs while reducing the design cost/effort compared to large System-on-Chips (SoCs). However, a major challenge in design of SiPs with many dies is automated design and insertion of Input/Output (I/O) cells to minimize energy and delay of the wire traces. This paper presents an automated cell library generation flow for all-digital I/O circuits for SiP integration. Given parameterized models of SiP wire traces, our method automatically designs, optimizes, and generates layouts of I/O cells for delay/energy minimization. The proposed flow is demonstrated on interposer based SiP integration considering 28nm CMOS technology and 65nm BEOL technology. Given a multi-die SiP design and associated interposer wire-traces, this paper demonstrates that automated I/O library cell generation can reduce maximum die-to-die communication delay or energy. We demonstrate the proposed flow for various interposer parameters and SiP designs to show the feasibility of chip-interposer co-design.