Responsive Banner

Optimasi Sinyal Respon Sistem Sensor Rasa dengan Metode Robust Locally Weighted Scatterplot Smooth (RLOESS) untuk Mereduksi Noise menggunakan Gui Matlab

Sugianto, Febi (2015) Optimasi Sinyal Respon Sistem Sensor Rasa dengan Metode Robust Locally Weighted Scatterplot Smooth (RLOESS) untuk Mereduksi Noise menggunakan Gui Matlab. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.

[img] Text (Fulltext)
10640019.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (3MB) | Request a copy

Abstract

ABSTRAK

Salah satu pengembangan sensor biomimetic adalah Lidah Elektronik (e-tongue) yang bermanfaat untuk mengenali berbagai jenis rasa dengan mendeteksi dan memonitoring evolusi senyawa kimia. Pola sinyal yang diperoleh dari gugusan sensor (sensor array) pada Lidah Elektronik (e-tongue) masih didominasi oleh pengaruh noise maupun outlier sehingga sulit untuk mengetahui pola respon sinyal dari data. Penelitian ini bertujuan untuk mengoptimasi sinyal respon dinamik sensor rasa, sehingga noise dapat tereduksi dan juga untuk menggambarkan suatu fungsi deterministik dari variasi dalam data, dengan fitting sederhana sejumlah set data sampel. Metode optimasi yang diterapkan diantaranya adalah Moving Average, Locally Weighted Scatterplot Smooth (LOESS), dan Robust Locally Weighted Scatterplot Smooth (RLOESS). Data yang digunakan merupakan data yang diambil dari hasil penelitian sensor rasa pada Lidah Elektronik (e-tongue), berupa data sinyal respon sistem sensor rasa. Hasil penelitian menunjukkan Robust Locally Weighted Scatterplot Smooth (RLOESS) memiliki kestabilan dan fleksibilitas antara titik data respon, reduksi noise yang signifikan dan ketahanan terhadap outlier lebih baik dibandingkan dengan Moving Average, sebagai penguatan dari Locally Weighted Scatterplot Smooth (LOESS). Pengurangan rata-rata dari titik data atau sample point terhadap titik-titik data asli adalah sekitar 1,967259.

ABSTRACT

One of the biomimetic sensors development is an Electronic Tongue (e-tongue) which is useful to identify different types of flavors by detecting and monitoring the evolution of chemical compounds. Signal pattern obtained from a sensor arrays on Electronic Tongue (e-tongue) is still dominated by the influence of noise and outliers, therefore it is difficult to know the response pattern of the data signal. This research aims to optimize the dynamic response of the sensor signal flavor, then noise can be reduced and also to describe a deterministic function of the variation in the data, with a simple fitting from a set number of data samples. Optimization methods applied include Moving Average, Locally Weighted Scatterplot Smooth (LOESS), and Robust Locally Weighted Scatterplot Smooth (RLOESS). The data used is the data that is retrieved from the taste sensor research on Electronic Tongue (e-tongue), in the form of the response signal data of taste sensing system. The results showed Robust Locally Weighted Scatterplot Smooth (RLOESS) have stability and flexibility of response data points, the significant of noise reduction and resistance to outliers better than the Moving Average, as the strengthening of Locally Weighted Scatterplot Smooth (LOESS). The Average reduction of data points or sample point to the original data points is approximately 1,967259.

Item Type: Thesis (Undergraduate)
Supervisor: Tazi, Imam and Abtokhi, Ahmad
Contributors:
ContributionNameEmail
UNSPECIFIEDTazi, ImamUNSPECIFIED
UNSPECIFIEDAbtokhi, AhmadUNSPECIFIED
Keywords: Lidah Elektronik (E-Tongue); Moving Average; Locally Weighted Scatterplot Smooth (LOESS); Robust Locally Weighted Scatterplot Smooth (RLOESS); Pola Respon Sinyal Electronic Tongue (E-Tongue); Moving Average; Locally Weighted Scatterplot Smooth (LOESS); Robust Locally Weighted Scatterplot Smooth (RLOESS); Signal Response Pattern
Departement: Fakultas Sains dan Teknologi > Jurusan Fisika
Depositing User: Moch. Nanda Indra Lexmana
Date Deposited: 03 May 2023 14:01
Last Modified: 03 May 2023 14:01
URI: http://etheses.uin-malang.ac.id/id/eprint/49496

Downloads

Downloads per month over past year

Actions (login required)

View Item View Item