GSTF Journal on Computing (JoC)

, 4:9

First online:

Open Access This content is freely available online to anyone, anywhere at any time.

Blind Signal Separation for Medical Data Recording Using Self-Organizing Neural Network

  • Alvin SahroniAffiliated withYogyakarta-Bandung Institute of Technology, Gadjah Mada University
  • , Hendra SetiawanAffiliated withBandung Institute of Technology, Gadjah Mada University


This paper present a study of signal processing in Blind Source Separation (BSS) application for medical field, especially during medical data recording. There are two main techniques that will be investigated; Natural Gradient Method (NGM) and Self-Organized Neural Network (SONN). The main source signals are Electrocardiograph (ECG), and Electroencephalograph (EEG) that linearly mixed with noise signal. This study related to doctor’s assistant for remote outpatient and their patients to communicate each other in separate places, as well as the doctor will be able to monitoring patient’s condition using a telephone/mobile phone that have been attached with an embedded system. Hopefully, while implementing BSS technique into an embedded system for signal processing application will increase the quality of medical outpatient system remotely. This investigation conclude that SONN provide about 80% effectively than NGM while separating mixed source signals of EEG, ECG, and noise Signal.


BSS Neural Network EEG ECG Neural Network Signal Processing Computation