MATLAB代写-EEE 6209

Coursework 2020/21
Dr Charith Abhayaratne

1
EEE 6209
Coursework 2020-21: EEE6209
Dr Charith Abhayaratne
15/12/2020

Total Marks: 100
Marks to meet the threshold: 50
The coursework contributes to 40% of the grading assessment

TITLE: Noise Removal

Due Date: 21
st
April 2021 at 16.00 via Blackboard EEE6209 coursework submission link

The aim of the coursework exercise is to explore and apply the signal processing you learned in this module
for denoising a signal. Each of you has been assigned an individual signal. The number of the signal
“signals” file from Blackboard coursework folder and choose the MAT file (x.mat, where x is your signal
number). Also download the noise file (noise.MAT) from Blackboard coursework folder. Two variables
can be found in the signal MAT file. They are the signal assigned to you and the signal number, n. The
noise MAT file contains the noise signal (V) to corrupt the signal assigned to you.

Add noise, V, to your test signal, A, to get the corrupted image, B, as follows:
B= A+ V.

You have to complete the following 4 tasks:
1. Analyse the original signal A, to understand its time-domain and frequency domain characteristics.
2. Use the Fourier transform-based techniques to remove the noise in signal, B, to recover the original
signal, A.
3. Explore the Moving Average Filter (MAF) you learned in lectures (Topic 02) to remove noise in
signal, B, to recover the original signal, A.
4. Use the Discrete Cosine Transform (DCT) to remove noise from B to recover the original signal 4

You are required to submit your findings in the following two assessments. For each task in the
assessments 1 and 2, you are required to do the following sub tasks and provide your answers in the
submission. Do not copy text from the task descriptions. Just use the relevant sub task number (e.g., 1.1.1,
1.2.1, etc.)

Assessment 1 (Description of the methodology, discussion of results and conclusions) Not more than 2000
words

1.1.1 Signal ID
1.1.2 The methodology you used to determine time-domain and frequency domain signal characteristics.
(Hint: Firstly, determine the time domain characteristics and use them to segment the signal A into
smaller segments with similar time-domain characteristics. The determine the frequency domain
characteristics for each of the segments)
1.1.3 Discussion of the time domain and frequency domain characteristics of the signal A
1.1.4 Based on 1.1.2 and 1.1.3 explain how you would choose smoothing filters with different parameters
and thresholds for transform domain noise removal for the signal B.

Coursework 2020/21
Dr Charith Abhayaratne

2
1.2.1 The methodology for the Fourier transform-based noise removal. If your technique uses any
parameter values, for example, threshold values, you need to explain how they were chosen.
1.2.2 Discuss the performance of the Fourier transform –based noise removal algorithm and the choice of
its parameter values based on your response to 2.2.1 and 2.2.2.

1.3.1 The methodology for the MAF-based noise removal using the M-point MAF. If your technique uses
any parameter values, for example, filter length (M) and the number of passes, you need to explain how
they were chosen.
1.3.2 Discuss the performance of the MAF–based noise removal algorithm and the choice of its parameter
values, such as, filter length (M) and the number of passes based on your response to 2.3.1 and 2.3.2.

1.4.1 The methodology for the DCT-based noise removal using the N-point DCT. If your technique uses
any parameter values, for example, threshold values and the size (N) of DCT, you need to explain how they
were chosen.
1.4.2 Discuss the performance of the DCT –based noise removal algorithm and the choice of its parameter
values, such as, threshold values and the size (N) of DCT based on your response to 2.4.1 and 2.4.2.

1.5.1 Provide your conclusions based on the summary table in 2.5.1 and any suggestions for further
improvements.

Assessment 2 (Numerical results and plots) – No page/word limit

2.1.1 Signal ID
2.1.2 Show the plots of original and noisy signals
2.1.3 Show plots to support the discussion of the time domain characteristics in 1.1.2 and 1.1.3
2.1.4 Show plots to support the discussion of the frequency domain characteristics in 1.1.2 and 1.1.3
2.1.5 Show the mean square error (MSE) values for the segments of noisy signal B with respect to those of
signal A.

2.2.1 Compute the MSE values for the recovered signal segments in 1.2.1 with respect to those of signal A
for different choices of parameters you have used.
2.2.2 Plot the de-noised signal segments in 1.2.1.

2.3.1 Compute the MSE values for the recovered signal segments in 1.3.1 with respect to those of signal A
for different choices of parameters you have used.
2.3.2 Plot the de-noised signal segments in 1.3.1.

2.4.1 Compute the MSE values for the recovered signal segments in 1.4.1 with respect to those of signal A
for different choices of parameters you have used.
2.4.2 Plot the de-noised signal segments in 1.4.1.

2.5.1. Provide a table summarising the results (MSE for each segment) from above three methods.

The marking scheme is as follows:

Assessment 1:
1.1.1-1.1.3 10 marks
1.1.4 10 marks
1.2.1 10 marks
1.2.2 10 marks
1.3.1 15 marks
Coursework 2020/21
Dr Charith Abhayaratne

3
1.3.2 15 marks
1.4.1 10 marks
1.4.2 10 marks
1.5.1 10 marks

Assessment 2:
2.1.1-2.1.5 (accuracy) 20 marks
2.4.1-2.2.2 (accuracy) 20 marks
2.4.1-2.3.2 (accuracy) 30 marks
2.4.1-2.4.2 (accuracy) 20 marks
Quality of figures/tables 10 marks

Overall marks = 0.7x(Assessment 1) + 0.3x(Assessment 2)

- MSE definition

,

where L is the signal length.

- DO NOT provide Matlab codes with your report. Instead use pseudo codes to explain your
methodology and algorithms.
- DO NOT copy and paste any text from this document. Doing so will result in a matching in
Turnitin reports leading to deduction of marks by the department’s plagiarism assessment
committee. Therefore just use the question numbers, 1.1.3, 1.1.2,….. etc
- All your numerical answers should include the appropriate units. All the plots should include x-axis
label, y-axis label, plot title and a legend if you are showing multiple plots in the same figure. Not
following these instructions may result in penalty marks.