Enhancing Tsunami Warnings with Low-Rank Matrix Estimation for Signal Reconstruction

Dhenok Ayu Setianingsih*, Dhany Arifianto, Suyanto, Margiasih Putri Liana

*Corresponding author for this work

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

Abstract

This study aims to improve signal reconstruction for tsunami early warning systems by utilizing low-rank matrix estimation to handle sparse underwater acoustic data. The objectives are to measure signal sparsity using the Gini Index and to compare the effectiveness of three reconstruction methods: Orthogonal Least Square (OLS), Orthogonal Matching Pursuit (OMP), and L1 Norm. The methods involve processing acoustic data from underwater experiments and evaluating reconstruction success based on error magnitude, computation time, and cosine similarity. Results indicate that the L1 Norm method outperforms OLS and OMP in terms of lower error rates and faster computation times. The findings have significant implications for enhancing the reliability and timeliness of tsunami early warnings, with potential applications in other fields such as medical imaging and wireless communications.

Original languageEnglish
Title of host publication2024 3rd International Conference on Creative Communication and Innovative Technology, ICCIT 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350367492
DOIs
Publication statusPublished - 2024
Event3rd International Conference on Creative Communication and Innovative Technology, ICCIT 2024 - Hybrid, Tangerang, Indonesia
Duration: 7 Aug 20248 Aug 2024

Publication series

Name2024 3rd International Conference on Creative Communication and Innovative Technology, ICCIT 2024

Conference

Conference3rd International Conference on Creative Communication and Innovative Technology, ICCIT 2024
Country/TerritoryIndonesia
CityHybrid, Tangerang
Period7/08/248/08/24

Keywords

  • L1 Norm
  • OLS
  • OMP
  • low rank matrix
  • reconstruction

Fingerprint

Dive into the research topics of 'Enhancing Tsunami Warnings with Low-Rank Matrix Estimation for Signal Reconstruction'. Together they form a unique fingerprint.

Cite this