TY - JOUR
T1 - PINSim
T2 - A Processing In- and Near-Sensor Simulator to Model Intelligent Vision Sensors
AU - Tabrizchi, Sepehr
AU - Morsali, Mehrdad
AU - Pan, David
AU - Angizi, Shaahin
AU - Roohi, Arman
N1 - Publisher Copyright:
© 2002-2011 IEEE.
PY - 2024
Y1 - 2024
N2 - This letter introduces PINSim, a user-friendly and flexible framework for simulating emerging smart vision sensors in the early design stages. PINSim enables the realization of integrated sensing and processing near and in the sensor, effectively addressing challenges such as data movement and power-hungry analog-to-digital converters. The framework offers a flexible interface and a wide range of design options for customizing the efficiency and accuracy of processing-near/in-sensor-based accelerators using a hierarchical structure. Its organization spans from the device level upward to the algorithm level. PINSim realizes instruction-accurate evaluation of circuit-level performance metrics. PINSim achieves over $25,000\times$ speed-up compared to SPICE simulation with less than a $4.1\%$ error rate on average. Furthermore, it supports both multilayer perceptron (MLP) and convolutional neural network (CNN) models, with limitations determined by IoT budget constraints. By facilitating the exploration and optimization of various design parameters, PiNSim empowers researchers and engineers to develop energy-efficient and high-performance smart vision sensors for a wide range of applications.
AB - This letter introduces PINSim, a user-friendly and flexible framework for simulating emerging smart vision sensors in the early design stages. PINSim enables the realization of integrated sensing and processing near and in the sensor, effectively addressing challenges such as data movement and power-hungry analog-to-digital converters. The framework offers a flexible interface and a wide range of design options for customizing the efficiency and accuracy of processing-near/in-sensor-based accelerators using a hierarchical structure. Its organization spans from the device level upward to the algorithm level. PINSim realizes instruction-accurate evaluation of circuit-level performance metrics. PINSim achieves over $25,000\times$ speed-up compared to SPICE simulation with less than a $4.1\%$ error rate on average. Furthermore, it supports both multilayer perceptron (MLP) and convolutional neural network (CNN) models, with limitations determined by IoT budget constraints. By facilitating the exploration and optimization of various design parameters, PiNSim empowers researchers and engineers to develop energy-efficient and high-performance smart vision sensors for a wide range of applications.
KW - neural network
KW - numerical simulation
KW - processing-in-sensor
KW - processing-near-sensor
UR - http://www.scopus.com/inward/record.url?scp=85213430907&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85213430907&partnerID=8YFLogxK
U2 - 10.1109/LCA.2024.3522777
DO - 10.1109/LCA.2024.3522777
M3 - Article
AN - SCOPUS:85213430907
SN - 1556-6056
JO - IEEE Computer Architecture Letters
JF - IEEE Computer Architecture Letters
ER -