Andrew Berkovich

Andrew Berkovich

Auto-generated profile, based on publicly available information

Andrew Berkovich

Reality Labs Research, Meta Inc., Redmond, WA, USA

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 January 2022 in IEEE Transactions on Electron Devices
Verified
Journal Article
Closed access
RIRimon IKENO
Rimon IKENO
KMKazuya Mori
Kazuya Mori
MUMasayuki Uno
Masayuki Uno
KMKen Miyauchi
Ken Miyauchi
TIToshiyuki Isozaki
Toshiyuki Isozaki
CLChiao Liu
Chiao Liu
This article presents a global shutter (GS), stacked digital pixel sensor (DPS) to meet the ultra-low power, ultra high dynamic range (HDR) requirements for battery-powered, always-on mobile computer vision (CV) applications. The $4.6~\mu \text{m}$ DPS pixel is partitioned into two layers; a dual conversion gain (CG) pixel with a backside-illuminated photodiode (PD) on the top layer and an in-pixel ADC circuit with 10 bit SRAM on the bottom layer. A Cu-to-Cu hybrid-bonding (HB) technology is used to connect the two layers via pixel-level interconnect. The sensor features an overlapped triple quantization (3Q) scheme that performs a time-to-saturation (TTS) quantization and the dual-CG linear ADC mode sequentially in the same frame and extends the dynamic range (DR) with a small number of ADC bits. The DPS with a $512\times512$ pixel array has achieved an ultra-HDR of 127 dB and low power consumption of 5.75 mW at 30 frames per second. The nonlinear conversion characteristics of the TTS mode provide an equivalent full well capacity (FWC) of 9000 ke−, while the high CG linear ADC mode realizes a low noise floor of 4.2 e−.

1 December 2021 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Verified
Journal Article
Closed access
RPReid Pinkham
Reid Pinkham
ABAndrew Berkovich
Andrew Berkovich
ZZZhengya Zhang
Zhengya Zhang
Untethered Augmented and Virtual Reality (AR/VR) devices are an emerging compute platform with unique opportunities and challenges. AR/VR devices use an array of sensors, including multiple cameras, to understand their surroundings and provide the user with an immersive experience. To deliver the functionality and performance, AR/VR devices rely on state-of-the-art algorithms including Deep Neural Networks (DNNs). These algorithms must operate in real time, and it presents a computational challenge for a mobile system. The emergence of on-sensor compute provides a possible solution to increase the processing capabilities of an AR/VR platform. In this work, we explore how to optimally map DNN models on an AR/VR compute platform that consists of an on-sensor processor and an edge processor to minimize energy and latency. We explore properties of popular DNN models, and the ideal network split locations, processor sizes, caching strategies, and the interactions between these design choices using a new Distributed Algorithm Simulator (DAS). Based on this study, we develop the basic principles on network split, parameter caching, and two-processor balancing to achieve near-optimal system designs. We show the addition of on-sensor processing to the existing Quest 2 VR platform can reduce MobileNetV3 inference energy by 64.6%. Finally, we demonstrate in a representative AR/VR platform how the minimum-energy configuration changes under the practical design constraints of memory size and silicon area, as well as the impact of future memory technologies.

25 April 2022 in IEEE Transactions on Electron Devices
Verified
Journal Article
Open access
KMKazuya Mori
Kazuya Mori
NYNaoto Yasuda
Naoto Yasuda
KMKen Miyauchi
Ken Miyauchi
TIToshiyuki Isozaki
Toshiyuki Isozaki
ITIsao Takayanagi
Isao Takayanagi
CLChiao Liu
Chiao Liu
It is anticipated that ubiquitous computer vision (CV) and artificial intelligence (AI) applications used on mobile devices will grow significantly. Such applications require battery-powered, always-on mobile devices to support indoor/outdoor, day/night usages. A global shutter (GS), stacked digital pixel sensor (DPS) is a promising candidate to meet such requirements because of its potential for ultralow-power, ultrahigh dynamic range (HDR), and a small form factor. This article presents a prototype 4.0- ${\mu } \text{m}$ stacked DPS operating in its dual quantization (2Q) to realize HDR. The 4.0- ${\mu }\text{m}$ DPS pixel is formed with two layers, a backside illuminated pinned photodiode (PD) pixel on the top layer and an in-pixel analog-to-digital conversion (ADC) circuit with 9-bit static random access memory (SRAM) on the bottom layer. A Cu-to-Cu hybrid bonding (HB) technology is used to connect the two layers via pixel-level interconnect. In the 2Q scheme, a time-stamp (TS) quantization and a linear ADC are performed sequentially in the same frame, which extends the dynamic range (DR) with a small number of ADC bits of 9. The DPS with a 1024 $\times$ 832 pixel array has achieved a single-exposure ultra HDR of 107 dB in a single frame. The nonlinear conversion characteristic of the TS mode provides an equivalent full well capacity (FWC) of 2000ke−, while the noise floor in the linear ADC mode is 8.3e−.

23 December 2022 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
RPReid Pinkham
Reid Pinkham
JEJack Erhardt
Jack Erhardt
BSBarbara De Salvo
Barbara De Salvo
ABAndrew Berkovich
Andrew Berkovich
ZZZhengya Zhang
Zhengya Zhang
Advanced edge sensing/computing devices, such as AR/VR devices, have a uniquely challenging adaptive baseline workload and camera sensor structure. These devices must process images in real-time from multiple sensors, placing a large burden on a typical centralized mobile SoC processor. Augmenting the sensors with a package-integrated near-sensor processor can improve the device’s processing performance as well as reduce energy consumption. This near-sensor processor must adapt to the dynamic workloads, fit within a limited silicon footprint and energy envelope, and satisfy the real-time requirement. In this work, we present ANSA, a near-sensor processor architecture supporting flexible processing schemes and dataflows to maintain high efficiency for dynamic CNN workloads. ANSA is scalable to sub-mm2 sizes to match the footprint of advanced image sensors. ANSA supports module-level power gating to adapt the compute capacity to dynamic workloads. Finally, ANSA leverages recent advancements in high-density non-volatile memory and 3D packaging to support weight storage within the area constraints of an image sensor. Overall, ANSA achieves inference energy consumption up to $30\times $ lower than a standard SIMD baseline. Additionally, our design’s scalability allows it to achieve up to $2.76\times $ lower average inference energy at $4.5\times $ lower silicon area compared to competing edge accelerator designs.

1 December 2011
Verified
Conference Paper
Closed access
SSSaeed Esmaili Sardari
Saeed Esmaili Sardari
ABAndrew Berkovich
Andrew Berkovich
AIAgis A. Iliadis
Agis A. Iliadis
A study of the conductivity type of undoped high quality ZnO thin-films grown on (100) Si substrates by pulsed laser deposition (PLD), is reported. Two groups of samples were prepared at a wide range of growth temperatures between 150 and 350°C and oxygen overpressure of 5.5×10 -6 Torr (low oxygen pressure regime) for one group and 2.5×10 -4 Torr (high oxygen pressure regime) for the other group. The electrical properties of the samples were studied by Hall Effect measurements using the four-point van der Pauw method. The measurements were performed using an Ecopia HMS-5000 Hall Effect Measurement System, with fixed magneto-flux density of 0.51T, and injected current that can be controlled between 1nA to 20mA. The system allows for variable temperature readings between 77 and 350K. Standard photolithography was used to prepare a cloverleaf mesa pattern of ZnO for the Hall Effect measurements.