Temperature compensation for humidity sensors using ISSA-BP neural network

Dapeng Li, Hechu Zhang, Yu Yang, Wei Li, Shuai Wang, Dezhi Zheng*

*Corresponding author for this work

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

Abstract

High-precision humidity detection is essential in various fields. However, the sensor's output signal is often affected by complex environmental temperature changes. To address these challenges, this paper propose an improved sparrow search algorithm based on the back propagation neural network (ISSA-BP). This method optimizes the initialization process of the traditional algorithm and introduces an edge evolution strategy to enhance its iterative update process, significantly improving both efficiency and accuracy. Experimental results demonstrate that the proposed algorithm reduces the mean absolute percentage error (MAPE) from 4.35% to 0.92% and decreases the convergence time from 1.25 s to 0.65 s, compared to traditional methods.

Original languageEnglish
Title of host publicationACM MobiCom 2024 - Proceedings of the 30th International Conference on Mobile Computing and Networking
PublisherAssociation for Computing Machinery, Inc
Pages2004-2009
Number of pages6
ISBN (Electronic)9798400704895
DOIs
Publication statusPublished - 29 May 2024
Event30th International Conference on Mobile Computing and Networking, ACM MobiCom 2024 - Washington, United States
Duration: 18 Nov 202422 Nov 2024

Publication series

NameACM MobiCom 2024 - Proceedings of the 30th International Conference on Mobile Computing and Networking

Conference

Conference30th International Conference on Mobile Computing and Networking, ACM MobiCom 2024
Country/TerritoryUnited States
CityWashington
Period18/11/2422/11/24

Keywords

  • back propagation neural network
  • humidity sensor
  • improved sparrow search algorithm
  • temperature compensation

Fingerprint

Dive into the research topics of 'Temperature compensation for humidity sensors using ISSA-BP neural network'. Together they form a unique fingerprint.

Cite this