Effect of ADC to respiratory sound classification

• Main Author: Lau, Benson Yong Zhi
• Other Authors: Ser Wee
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/139787
• Institution: Nanyang Technological University

Excessive amount of water in the lung is a common symptom for pulmonary edema which could result in heart failure. A diagnostic method using the acoustic technique has been proposed. For this to work, the diaphragm will be connected to handphones which consist of a diagnostic mobile application. The mobile application will process the audio signal and classify it accordingly. However, in the modern world, there are lots of mobile makers which include some of the big brands like Huawei, Samsung, Apple, LG. Some of these companies produced their own chips. Hence this results in different phones having different specifications. This project proposed a study in the effect of Analog-Digital Converter in the classification of sound. The lung recording was provided by the master student who is working on the project. The sound classification algorithms will be developed which includes several steps such as signal preprocessing, feature extraction, feature selection, classification, and performance evaluation. The signal will then be undergoing the ADC algorithm before sending it to the trained classifier to study the effect of ADC on sound classification. Different ADC resolution, 4-bit, 8-bit, and 16-bit will be used to generate the features for classification on the trained model. From the findings obtained by the project, the author observed that both 8-bit and 16-bit resolution deviates closely similar while 4-bit resolution deviates very differently as compared to the other 2 resolution.