The IEEE International Symposium on Circuits and Systems (ISCAS 2022) will be held at Austin, Texas (USA) from May 28th to June 1st.
This event is the flagship conference of the IEEE Circuits and Systems Society and the world’s premiere forum for researchers in the highly active fields of theory, design and implementation of circuits and systems. ISCAS 2022 will be driven by the theme Emerging issues in intelligent, ubiquitous, autonomous, mobile devices to promote multidisciplinary solutions for the societal and engineering challenges of our times.
Two papers in the frame of the StorAIge project will be presented as lecture by ST Italy and Politecnico di Torino.
- Privacy-preserving Social Distance Monitoring on Microcontrollers with Low-Resolution Infrared Sensors and CNNs
Chen Xie, Francesco Daghero, Yukai Chen, Marco Castellano, Luca Gandolfi, Andrea Calimera, Enrico Macii, Massimo Poncino, Daniele Jahier Pagliari
Abstract : Low-resolution infrared (IR) array sensors offer a low-cost, low-power, and privacy-preserving alternative to optical cameras and smartphones/wearables for social distance monitoring in indoor spaces, permitting the recognition of basic shapes, without revealing the personal details of individuals. In this work, we demonstrate that an accurate detection of social distance violations can be achieved processing the raw output of a 8×8 IR array sensor with a small-sized Convolutional Neural Network (CNN). Furthermore, the CNN can be executed directly on a Microcontroller (MCU)-based sensor node. With results on a newly collected open dataset, we show that our best CNN achieves 86.3% balanced accuracy, significantly outperforming the 61% achieved by a state-of-the-art deterministic algorithm. Changing the architectural parameters of the CNN, we obtain a rich Pareto set of models, spanning 70.5-86.3% accuracy and 0.18-75k parameters. Deployed on a STM32L476RG MCU, these models have a latency of 0.73-5.33ms, with an energy consumption per inference of 9.86-68.57muJ.
- Phase-Change Memory in Neural Network Layers with Measurements-based Device Models
Carmine Paolino, Alessio Antolini, Fabio Pareschi, Mauro Mangia, Riccardo Rovatti, Eleonora Franchi Scarselli, Gianluca Setti, Roberto Canegallo, Marcella Carissimi, Marco Pasotti
Abstract: In this work, we describe a methodology that, starting from measurements performed on a set of real PCM devices, enables the training of a neural network. The core of the procedure is the creation of a computational model, sufficiently general to include the effect of unwanted nonidealities. Results show that, depending on the task at hand, a different level of accuracy is required in the PCM model applied at train-time to match the performance of a traditional, reference network. Moreover, the trained networks are robust to the perturbation of the weights.
More information on ISCAS 2022: https://www.iscas2022.org/
