Part A: Laboratory coursework The main purpose of the assignment is to demonstrate understanding and critical evaluation of a practical digital signal processing system. The assignment evaluates the critical understanding of analogue-to-digital conversion (ADC) and digital-to-analogue conversion (DAC) processes, distortion in a real system and techniques to reduce the noise. The aim of the course is to simulate a real-time digital signal processing system in Matlab environment, demonstrate understanding of a digital signal processing and critically evaluate the different blocks. Figure 1: A real-time digital signal processing system The assignment consists of following components: Part Assignment Details Marks A i) ii) iii) Consider a voice signal with maximum frequency of 4 kHz and the maximum voltage input of ±0.5V. The system consists of 8-bit DAC and ADC codec with sampling frequency of 40 kHZ. Write a Matlab program to represent the discrete time signal and display the quantized output. Assuming 4-bit, 8-bit and 16-bit ADC, demonstrate the quantisation errors due to bit resolutions. Display histogram of quantisation noises and calculate quantisation signal-to-noise ratio. Based on the quantisation noises and the system complexity, comment on the ADC you would practically implement. Demonstrate the aliasing effect by sampling the input signal at the sampling rates of 6, 12, 20 and 40 kHz and displaying results in appropriate time or frequency domain. 5% 10% 10% B i) ii) iii) iv) v) vi) Design a suitable linear phase filter to remove out-of-band noise if signal of interest has frequency between 1-4 kHz. Use filter design and analysis (FDA) toolbox in Matlab. Analysed the frequency response of the filter. Plot amplitude and phase responses. Verify the amplitude response of the filter by inputting signals with varying frequency sinusoidal signals to the filter. Does the amplitude match with the response in B(ii)? Assuming the signal is corrupted by additive white Gaussian noise, investigate the effect of noise on the signal with different signal-to-noise ratio (SNR). Demonstrate the effect of noise both in time and frequency domains. (Show at least three signals with SNRs in the range 1-15 dB). Develop a methodology to remove the noise from the signal. Compare the signal with and without noise removal for all the cases you considered in B(iv). Reconstruct the analogue signal after noise removal. Comment on the fidelity of recovered signal. 5% 5% 5% 10% 15% 5% C Write a concise report (marks will be deducted for unnecessary information) covering the details in Parts A – B. Keep the format of this similar as to that 10% you would use in your Final Year Project dissertation. Part B: Research Objectives: to enforce research based learning and to familiarise yourself with latest technologies. 1. Provide a literature review of wavelet transform based image denoising and compression. The report should include an introduction to wavelet transform, algorithms for image denoising and compression using wavelet, advantages of the wavelet technique over other techniques and Conclusion. The report should not be longer than 2-page A4 size paper including reference with times new roman font of size 10.