Novita, Andri (2025) Deteksi Pola Pernapasan menggunakan Radar Continuous Wave (CW) dengan Teknik Pengurangan Motion Artifact menggunakan Empirical Mode Decomposition (EMD) dan Continuous Wavelet Transform (CWT). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5023201016-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5023201016-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy |
Abstract
Pemantauan tanda vital seperti pola pernapasan secara non-kontak penting dilakukan untuk mendeteksi kondisi kesehatan secara dini tanpa menimbulkan ketidaknyamanan atau risiko infeksi. Penelitian ini mengembangkan sistem deteksi pola pernapasan menggunakan radar Continuous Wave (CW) yang dilengkapi dengan algoritma pengolahan sinyal berbasis Empirical Mode Decomposition (EMD) dan Continuous Wavelet Transform (CWT). Fokus utama dari sistem ini adalah reduksi motion artifact yang sering mengganggu akurasi deteksi sinyal pernapasan. Radar CW dipilih karena mampu merekam pergerakan mikro permukaan dada akibat respirasi secara presisi tanpa perlu kontak langsung. Sinyal pernapasan yang tertangkap diproses menggunakan EMD untuk memisahkan komponen respirasi dari motion artifact, kemudian dianalisis dalam domain waktu-frekuensi menggunakan CWT guna mengekstraksi fitur-fitur yang relevan secara fisiologis. Implementasi EMD berhasil menurunkan nilai Mean Absolute Error (MAE) dari 13,08 BPM menjadi 3,66 BPM, serta meningkatkan korelasi terhadap respiration belt dari 0,030 menjadi 0,822. Model Random Forest digunakan untuk mengklasifikasikan pola pernapasan ke dalam dua kategori, yakni pernapasan normal dan abnormal (tachypnea), dengan akurasi mencapai 83,75% menggunakan cross-validation dan 85% pada skema train-test split. Sistem ini menunjukkan potensi sebagai solusi akurat, portabel, dan non-invasif untuk pemantauan pola pernapasan, serta berpeluang dikembangkan lebih lanjut untuk klasifikasi multi-pola respirasi secara real-time.
=====================================================================================================================================
Non-contact monitoring of vital signs such as breathing patterns is important to detect health conditions early without causing discomfort or risk of infection. This research develops a breathing pattern detection system using Continuous Wave (CW) radar equipped with Empirical Mode Decomposition (EMD) and Continuous Wavelet Transform (CWT) based signal processing algorithms. The main focus of this system is the reduction of motion artifacts that often interfere with the accuracy of respiratory signal detection. CW radar was chosen because it is able to record the micro-movement of the chest surface due to respiration precisely without the need for direct contact. The captured breathing signal is processed using EMD to separate the respiration component from motion artifacts, then analyzed in the time-frequency domain using CWT to extract physiologically relevant features. The implementation of EMD successfully reduced the Mean Absolute Error (MAE) value from 13.08 BPM to 3.66 BPM and increased the correlation to the respiration belt from 0.030 to 0.822. The Random Forest model was used to classify breathing patterns into two categories, namely normal and abnormal breathing (tachypnea), with an accuracy of 83.75% using cross-validation and 85% on a split train-test scheme. The system shows potential as an accurate, portable and non-invasive solution for breathing pattern monitoring, and has the opportunity to be further developed for real-time classification of multi-pattern respiration.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Radar CW, EMD, CWT, Motion Artifact, Pola Pernapasan, Random Forest, Tachypnea, Pemantauan Non-Kontak. CW Radar, EMD, CWT, Motion Artifact, Respiratory Pattern, Random Forest, Tachypnea, Non-Contact Monitoring |
| Subjects: | R Medicine > R Medicine (General) > R856.2 Medical instruments and apparatus. T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis |
| Depositing User: | Andri Novita |
| Date Deposited: | 05 Aug 2025 08:47 |
| Last Modified: | 05 Aug 2025 08:47 |
| URI: | http://repository.its.ac.id/id/eprint/125681 |
Actions (login required)
 |
View Item |