Status: Completed
Lead partner: Institute of Electrical Engineering, Slovak Academy of Sciences
Author: Jaromír Klarák
Description:
The NEURALSENS project makes available a software framework developed for modelling and simulation of sensing neural networks (SeNNs) with sensor-dependent weights.
The software was developed within the project:
Smart gas and temperature sensors with neural-network-based low-level in-sensor data processing capability
Project No.: 09I05-03-V02-00058
Software repository
The software is publicly available at:
https://github.com/jaro221/NEURALSENS_PNN
Description
The repository contains Python scripts for modelling neural-network-based sensor systems developed within the NEURALSENS project. The framework focuses on the simulation of sensing neural networks, where the response of individual sensing elements can be represented by mathematical functions, including polynomial sensor characteristics.
The current version includes:
SeNN.pySeNN_Theory.py
The software demonstrates how sensor responses can be integrated into neural-network-like architectures and used for modelling low-level in-sensor data processing.
Main purpose
The purpose of this software is to support the development of hard-coded sensing neural networks (SeNNs), where the behaviour of individual sensing elements contributes directly to the computation performed by the sensor array.
This approach is relevant for future smart gas and temperature sensors with low-level in-sensor data processing capability.
Authors
Software author:
Jaromír Klarák
Developed within the NEURALSENS consortium.
Funding acknowledgement
This work was funded by the European Union NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project NEURALSENS, No. 09I05-03-V02-00058.