Main Page | Courses | Research | Publications | VILAB | AISENSE

Hulya Yalcin: Courses
EHB 372E - Digital Signal Processing and Design Applications Laboratory


This course will treat a broad range of Digital Signal Processing (DSP) topics.
It will strengthen the student's understanding of the foundations of DSP, introduce
the students to major application areas: speech processing, image processing, and
array signal processing, and provide extensive hands-on design experience.
The course is composed of topical units designed to support laboratory experiments
and a design project. The Foundations unit is followed by Image Processing unit.


EHB-372E Syllabus

Experimental Setups & Weekly Homeworks by Hulya Yalcin

Week-1 : 1st Experimental Setup - Convolution
Introduction to the course. Learning how to code in Matlab environment.

Week-2: Experimenting with EEG signals
1) We first read a *.mat file (
five_eeg.mat ) that includes many samples of EEG signals from patients with epilepsy seisure into matlab environment.
2) If you'd like to lean more about this data, you can check out following paper.
EEG signals classification using the K-means clustering and a multilayer perceptron neural network model
3) We create our own matlab script for generating Gaussian noise and add Gaussian noise to one of the EEG signals.
4) Processing data in time domain: We use convolution to remove the noise from the signal.
5) Processing data in frequency domain: We use fft to remove the noise from the signal.
Cinderellas Tale

Week-3: We continued analysing EEG Signals
Unfortunately, we couldn't accomplish last weeks task on time.
So we continued our experimentation with EEG Signals.

Week-4: Let's move to 2-d world. How about convolution for 2-d data?
1) We will first take a picture of "you" and read it into matlab environment.
2) Then we'll add different kinds of noise onto your picture.
Namely, "salt and pepper" noise and "Gaussian" noise.
3) You will write your own matlab codes to filter different kinds of noise.
a) remove salt-and-pepper noise, b) remove Gaussian noise, c) how about using a "bilateral" filter?
4) At 16:30, we will have an in-class challenge.
I'll give you a picture with noise.
I won't tell you the type of the noise added to the picture.
You will use your matlab functions to remove the noise from the image.
You have to compare the performance of each filter in removing the noise.
The person who achieves the best performance will win an edible award!
details of 3rd Week's in-class assignment can be found in this link.
Cinderellas Tale

Week-5: Experimenting with 2D Fourier Transform and Searching for Cinderella
We will have lot's of fun this week!
TASK 1: Developing a user interface for filtering in frequency domain
TASK 2: Searching for Cinderella
details of 5th Week's in-class assignment can be found in this link.
Cinderellas Tale

Week-6: Studying Short Time Fourier Transform (STFT)
We will study Short Time Fourier Transform (STFT)
details of 6th Week's in-class assignment can be found in this link.
Cinderellas Tale

Week-7: Experimenting with FIR and IIR filters
We will study FIR and IIR filters this week.
FIR Filter Design
IIR Filter Design
details of 7th Week's in-class assignment can be found in this link.

Week-8: Studying Wavelets
We will study wavelets.
Haar Wavelets
Detecting baby spine in ultrasound images using wavelets.
Cinderellas Tale

Week-9: Studying Edge Detection and Segmentation Algorithms
This week, we'll study edge detection and segmentation algorithms in 2D images.
details of 9th Week's in-class assignment can be found in this link.
Cinderellas Tale

MATLAB TOOLBOXES suggested for EHB372E

Signal Processing Toolbox
DSP System Toolbox
Image Processing Toolbox
Computer Vision System Toolbox
Curve Fitting Toolbox
Neural Network Toolbox
Statistics and Machine Learning Toolbox
Robotics System Toolbox


EHB372E - Project Proposal Form (Word Format)

EHB372E - Project Proposal Form (Pdf Format)

Reference Books for the Course

Mark Fowler's Flipped Classrom Material

Prof. Mark Fowler has a wonderful website for Signals and Systems, as well as Digital Signal Processing. For each lecture, there are lecture notes, video lectures and presentations. If you yould like to review Signal Processing, then this it THE website!

Prof. Mark Fowler

Signals and Systems by Marks Fowler

Digital Signal Processing - 1

Digital Signal Processing - 2

Experimental Setups by İlker Bayram

Week-6 : The Short Time Fourier Transform
Week-7 : FIR Filter Design
Week-8 : IIR Filter Design
Week-10 : Haar Wavelets
Week-12 : Sparsity Regularization

Main Page | Courses | Research | Publications | VILAB | AISENSE
Electronics and Communication Engineering
Istanbul Technical University
34469, Maslak/Istanbul