Skip to main navigation Skip to search Skip to main content

A novel smart pressure and temperature sensor with AI

Jiguang Li*, Zhao Huang, Shuping Song

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper focuses on creating a new type of smart pressure sensor that uses advanced artificial intelligence (AI) to measure pressure and temperature simultaneously more precisely and reliably than traditional methods. The sensor works by detecting both pressure and temperature differences, which are converted into tiny electrical signals. These signals are then processed using AI to ensure accuracy. An Si-based device structure with thermopile was designed for this sensor. The model is simulated with Python. Long Short-Term Memory (LSTM) deep learning model is implemented to separate the pressure and temperature signal. The AI successfully separated the pressure signal from the temperature signal with minimal error, demonstrating that the neural network effectively learned this sensor's characteristics.

Original languageEnglish
Title of host publicationInternational Conference on Advanced Semiconductors and Communications, ICASC 2025
EditorsHuolin Huang, Hu Sheng
PublisherSPIE
ISBN (Electronic)9781510694989
ISBN (Print)9781510694972
DOIs
Publication statusPublished - 10 Sept 2025
EventInternational Conference on Advanced Semiconductors and Communications, ICASC 2025 - Dalian, China
Duration: 13 Jun 202515 Jun 2025

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13805
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceInternational Conference on Advanced Semiconductors and Communications, ICASC 2025
Country/TerritoryChina
CityDalian
Period13/06/2515/06/25

Keywords

  • AI
  • LSTM
  • Pressure
  • Sensor
  • Temperature

Fingerprint

Dive into the research topics of 'A novel smart pressure and temperature sensor with AI'. Together they form a unique fingerprint.

Cite this