Italy

**Please read the paper on the following link:**

https://ieeexplore.ieee.org/document/9177177

**Abstract:**

A 10nm DNN inference accelerator compresses model size with tabulation hash-based fine-grained weight sharing and increases 8b-compute density by 3.4× to 1.6 TOPS/mm2. The compressed model DNN implements lightweight hashing circuits to compress fully connected and recurrent neural networks. Optimized shared weight address generation reduces MUX tree area overhead by 40%. Runtime hash table generation and weight mapping circuits enable a peak energy efficiency of 9.0TOPS/W at 450mV, 25°C. A 128×-compressed 3-layer LSTM classifies TIMIT phonemes with 85.6% accuracy for a total energy of 14µJ/classification, with &lt;0.5% degradation in accuracy over an uncompressed network.

ESSCIRC ESSDERC 2021

A 9.0TOPS/W Hash-Based Deep Neural Network Accelerator Enabling 128× Model Compression in 10nm FinFET CMOS

deep learning asic

hashing

speech recognition

lstm

inference

deep neural networks

model compression

The aim of ESSCIRC and ESSDERC is to provide an annual European forum for the presentation and discussion of recent advances in solid-state devices and circuits.
The level of integration for system-on-chip design is rapidly increasing.

This is made available by advances in semiconductor technology. Therefore, more than ever before, a deeper interaction among technologists, device experts, IC designers, and system designers is necessary. ESSCIRC and ESSDERC are governed by a common Steering Committee and share Plenary Keynote Presentations and Joint Sessions bridging both device and circuit communities.

Highlights:
* 3 joint keynote presentations
* 3 keynote presentations ESSCIRC
* 3 keynote presentations ESSDERC
* Invited papers with overall coverage of all aspects of advanced devices and circuits
* Presentation of IEEE and ESSCIRC/ESSDERC Awards
* Educational Events

JOINT PLENARY talks live broadcast: September 13, 2021<br>
ESSCIRC and ESSDERC KEYNOTE talks live broadcast: September 15, 2021<br>
LIVE Q&A INTERACTIVE sessions for all Conference Technical sessions: September 14 – 16 – 17, 2021<br>
LIVE Q&A INTERACTIVE sessions for all Educational Events: September 20-21-22, 2021<br>
(on demand videos available from September 6 until November 30, 2021)

**Please see here for the [ESSCIRC Guide](https://drive.google.com/file/d/10dghzuv4aQBty4QfTIY9Q8hXyqsue8Le/view?usp=sharing), [ESSDERC Guide](https://drive.google.com/file/d/10dx9y2tm9cHPDa4upNUYjfXCf44aPry5/view?usp=sharing), and the [Program at a Glance](https://drive.google.com/file/d/10eU0OLTXEpiNzI5hkqv4mSr43T8cedPP/view?usp=sharing)**


To access ESSCIRC ESSDERC 2021, you must pre-register for the conference.  Instructions on how to access the event will be provided directly to attendees and included on the ESSCIRC ESSDERC conference website. [Register now](https://www.esscirc-essderc2021.org/).

Please register for the ESSCIRC conference to join the event.

The aim of ESSCIRC and ESSDERC is to provide an annual European forum for the presentation and discussion of recent advances in solid-state devices and circuits.
The level of integration for system-on-chip design is rapidly increasing.

This is made available by advances in semiconductor technology. Therefore, more than ever before, a deeper interaction among technologists, device experts, IC designers, and system designers is necessary. ESSCIRC and ESSDERC are governed by a common Steering Committee and share Plenary Keynote Presentations and Joint Sessions bridging both device and circuit communities.





**Please read the paper on the following link:**

https://ieeexplore.ieee.org/document/9177177

**Abstract:**

A 10nm DNN inference accelerator compresses model size with tabulation hash-based fine-grained weight sharing and increases 8b-compute density by 3.4× to 1.6 TOPS/mm2. The compressed model DNN implements lightweight hashing circuits to compress fully connected and recurrent neural networks. Optimized shared weight address generation reduces MUX tree area overhead by 40%. Runtime hash table generation and weight mapping circuits enable a peak energy efficiency of 9.0TOPS/W at 450mV, 25°C. A 128×-compressed 3-layer LSTM classifies TIMIT phonemes with 85.6% accuracy for a total energy of 14µJ/classification, with <0.5% degradation in accuracy over an uncompressed network.

technical paper

This letter presents a RISC-V System-on-Chip (SoC) with fully integrated switched-capacitor DC–DC converters, adaptive clock generators, mixed-precision floating-point vector accelerators, a 5-Gb/s serial memory interface, and an integrated power management unit (PMU) manufactured in 28-nm FD-SOI. The vector accelerator improves performance and energy per task on a matrix multiplication kernel by 15x and 13x, respectively, and end-to-end performance on machine learning and graph analytical workloads by 8x–12x. Inclusion of microarchitectural counters and fine spatial power-domain granularity facilitate the predictive power-management algorithms that reduce energy per task by 13%–22% compared to the baseline scalar processor. System level simulations of a range of SoC architectural variations with multiple cores and vector accelerators complement the silicon measurements.

Programmable Fine-Grained Power Management and System Analysis of RISC-V Vector Processors in 28-nm FD-SOI

**Please read the paper on the following link:**

https://1worldcontent.com/ess/9092.pdf

**Abstract:**

In this paper, a 4.3e- RMS low noise 4.6μm Time-of-Flight pixel based on charge domain with kTC noise removal is proposed to enhance the depth camera image quality. The pixel takes advantages of 6μm gradually doped epitaxial layer for 88.5% demodulation contrast at 200MHz and 18.5% QE at 940nm. Buried channel transfer gates are used to enable low capacitive switching and allowing best-in-class 1.4μW/pixel power consumption at 200MHz. The pixel is fully isolated due to deep trench isolation and pixel bulk is biased at several hundreds of mV for ultimate low power Time-of-Flight sensor and system. The paper describes the design of 0.54Mpix camera (672 x 804 pixels) implementing demodulation circuits robust against EMI and multi-devices interferences. The performance of an indirect Time-of-Flight system is demonstrated with 1.0% depth noise under 40kLux at 30fps.

4.6μm Low Power Indirect Time-of-Flight Pixel Achieving 88.5% Demodulation Contrast at 200MHz for 0.54MPix Depth Camera

**Please read the paper on the following link:**

https://1worldcontent.com/ess/9214.pdf

**Abstract:**

We will present major advances in Time of Flight image sensors for consumer applications. Several innovative elements integrated into a 130nm front side illumination CMOS node reveal superior sensor performance so far only attributed to much more complex backside illumination technologies. The gate controlled pixels comprising deep trenches, buried mirrors and integrated prisms, reveal high quantum efficiency and modulation transfer function close to the physical pixel size. In-pixel common mode suppression prevents saturation even under backlight conditions. The comprehensive System on Chip integrates high speed ADCs, a flexible phase shifter, as well as current monitoring for laser safety.

Advancements in Indirect Time of Flight Image Sensors in Front Side Illuminated CMOS

**Please read the paper on the following link:**

https://1worldcontent.com/ess/9072.pdf

**Abstract:**

Dark Count Rate (DCR) in Single-Photon Avalanche Diodes (SPAD) in Complementary Metal-Oxide Semiconductor technology is characterized and analyzed with a comprehensive simulation methodology. Based on a series of measurements of SPAD with various architectures, on an extended range of voltages and temperatures, the DCR measurements are correlated to the spatial localization of traps within the device and their parameters. To this aim, process and electrical simulations using Technology Computer-Aided Design (TCAD) tools are combined with an in-house McIntyre solver to compute the breakdown probability (Pt). The traps are accounted for using thermal SRH carrier generation-recombination mechanism which is coupled with the position-dependent breakdown probability. This rigorous methodology makes it possible to directly compare with DCR measurements, since only generated carriers with a non-negligible avalanche probability are considered.

Dark Count Rate in Single-Photon Avalanche Diodes: Characterization and Modeling Study

**Please read the paper on the following link:**

https://1worldcontent.com/ess/9005.pdf

**Abstract:**

A novel sun sensor concept is presented. Photodiodes operating as solar cells with continuous operation and dedicated logic to calculate the centroid position are integrated to achieve lower latency and energy consumption. The output data flow is remarkably reduced because the centroid of the illuminated pixels is the only sensor output data. It overcomes conventional digital sun sensors based on Active Pixel Sensor (APS) pixels and Address Event Representation (AER) in terms of latency and power consumption. Its latency is in the order of microseconds with an average power consumption lower than 100 uW. Experimental results are provided and benchmarked.

A High-Speed Low-Power Sun Sensor with Solar Cells and Continuous Operation

**Please read the paper on the following link:**

https://1worldcontent.com/ess/9212.pdf

**Abstract:**

The goal of this work is the design, realization, and characterization of a pixel-level front-end in 0.15μm CMOS technology, that allows to directly convert in the digital domain the weak output signal from the Field-Effect Transistor (FET) TeraHertz (THz) detector, in a noise-efficient way. The chosen architecture is a Continuous-Time Sigma-Delta Incremental Analog-to-Digital Converter (CT ΣΔ IADC) with a current DAC feedback, chopper technique and digital lock-in modulation. The measured SNR of this interface is 67.45dB, corresponding to an equivalent number of bit (ENOB) of 10.91. The pixel readout area is 0.072mm2, and the power consumption is 65μW from 1.8V supply. The Noise Equivalent Power (NEP) of the THz detector and readout chain is 268pW/√𝐇𝐳. The readout channel Input Referred Noise (IRN) is 1.2μVrms with the FET, and 0.94μVrms for the channel without detector.

A 0.94-μVrms Input Noise Pixel-Level Continuous Time ΣΔ IADC Interface for THz Sensing

**Please read the paper on the following link:**

https://ieeexplore.ieee.org/document/9146537

**Abstract:**

The dual-mode logic (DML) defines runtime adapted digital architectures that switch to either improved performance or lower energy consumption as a function of the actual computational workload. This flexibility is demonstrated for the first time by silicon measurements on a 16 × 16 -b Booth multiplier fabricated as a part of an ultralow-power digital signal processing (DSP) architecture for 16-nm FinFET technology. When running in the full-speed mode, the DML multiplier can achieve a performance boost of 19.5% as compared to the equivalent standard CMOS design. The same circuit saves precious energy (-27%, on average) when the energy-efficient mode is enabled, while occupying 13% less silicon area.

A 0.8-V, 1.54-pJ/940-MHz Dual-Mode Logic-Based 16×16-B Booth Multiplier in 16-nm FinFET

Multi-channel potentiostats are required for monitoring electrochemical processes in a time- and cost- efficient manner. State-of-the-art PCB potentiostats suffer from limited performance. Existing integrated CMOS potentiostats enable large number of parallel sensing channels, yet are still limited in the number of parallel stimulation channels and their bandwidth. The presented 40nm CMOS potentiostat chip overcomes this bottleneck with 96 individually controllable channels with 125dB current dynamic range and up to 150kHz bandwidth using an on-chip digital feedback controller. The chip supports both on-chip and off-chip electrodes allowing flexible electrode dimensions.

A 96-channel 40 nm CMOS Fully-Integrated Potentiostat for Electrochemical Monitoring

This keynote is describing the evolution of Artificial Intelligence (AI) adoption to the Edge for maximizing Data processing at the User experience level to treat as much as possible data analysis without saturating Internet traffic and Cloud data centers. Evolution of Neural Network (NN) computing through progressive usage of In Memory Computing (IMC) is also described. A few examples of Artificial Intelligence at the Edge are also commented. We then describe how at ST, we address competitive solutions for Edge AI thanks to differentiated FD-SOI CMOS technology, Phase Change Memory (PCM) embedded memory integration and advanced NN and IMC design solutions.

Downloads

Next from ESSCIRC ESSDERC 2021

Programmable Fine-Grained Power Management and System Analysis of RISC-V Vector Processors in 28-nm FD-SOI

Similar lecture

A 2.9-33.0 TOPS/W Reconfigurable 1D/2D Compute-Near-Memory Inference Accelerator in 10nm FinFET CMOS

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ESSCIRC ESSDERC 2021

Programmable Fine-Grained Power Management and System Analysis of RISC-V Vector Processors in 28-nm FD-SOI

Similar lecture

A 2.9-33.0 TOPS/W Reconfigurable 1D/2D Compute-Near-Memory Inference Accelerator in 10nm FinFET CMOS

Downloads