Course Title: Digital Signal Processing (Level Two)
Course Objectives:
1. Develop a deeper understanding of advanced digital signal processing techniques.
2. Learn advanced filter design methods, including multirate and adaptive filters.
3. Explore spectral analysis and estimation techniques.
4. Gain hands-on experience with DSP algorithms using software tools (e.g., MATLAB,
Python).
5. Apply advanced DSP techniques to real-world problems in audio, image, and
communication systems.
Course Outline:
Module 1: Review of Fundamental Concepts
- Recap of discrete-time signals and systems
- Review of Fourier transform, Z-transform, and discrete Fourier transform (DFT)
- Recap of basic digital filter design (FIR and IIR filters)
Module 2: Advanced Digital Filter Design
- *FIR Filter Design*:
- Windowed design method
- Frequency sampling method
- Optimal equiripple design (Parks-McClellan algorithm)
- IIR Filter Design:
- Bilinear transform method
- Impulse invariance method
- Comparison of FIR and IIR filters
- *Multirate Signal Processing*:
- Upsampling and downsampling
- Polyphase decomposition
- Applications in communication and audio systems
Module 3: Spectral Analysis and Estimation
- Nonparametric Methods:
- Periodogram and modified periodogram
- Welch's method
- Bartlett's method
- Parametric Methods:
- Autoregressive (AR), moving average (MA), and ARMA models
- Yule-Walker equations
- Levinson-Durbin recursion
- High-Resolution Spectral Estimation:
- MUSIC algorithm
- ESPRIT algorithm
Module 4: Adaptive Filtering
- Introduction to Adaptive Filters:
- Applications: noise cancellation, echo cancellation, system identification
- LMS Algorithm:
- Least mean squares (LMS) algorithm
- Convergence and stability analysis
- RLS Algorithm:
- Recursive least squares (RLS) algorithm
- Comparison with LMS
- Applications:
- Adaptive noise cancellation
- Adaptive beamforming
Module 5: Advanced Topics in DSP
- Wavelet Transform:
- Continuous and discrete wavelet transforms
- Applications in signal compression and denoising
- Time-Frequency Analysis:
- Short-time Fourier transform (STFT)
- Wigner-Ville distribution
- Applications in Image Processing:
- Image filtering and enhancement
- Edge detection and feature extraction
Module 6: DSP in Real-World Systems
- Audio Signal Processing:
- Speech processing and recognition
- Audio compression (e.g., MP3, AAC)
- Communication Systems:
- Digital modulation and demodulation
- Channel equalization
- *Biomedical Signal Processing*:
- ECG and EEG signal analysis
- Applications in healthcare
Laboratory/Software Sessions:
1. Advanced Filter Design:
- Designing and testing FIR and IIR filters using MATLAB/Python.
2. Spectral Analysis:
- Implementing periodogram, Welch's method, and parametric spectral estimation.
3. Adaptive Filtering:
- Implementing LMS and RLS algorithms for noise cancellation.
4. Wavelet Transform:
- Applying wavelet transforms for signal denoising and compression.
5. Applications:
- Implementing DSP techniques for audio, image, and communication systems.
Assessment Methods:
1. Quizzes and Assignments:
- Problem-solving exercises and coding tasks.
2. Midterm Exam:
- Covers Modules 1–3 (review, advanced filter design, spectral analysis).
3. Final Exam:
- Comprehensive exam covering all modules.
4. Lab Reports:
- Documentation of experiments and results.
5. Project:
- A small DSP project (e.g., audio noise cancellation, image enhancement).
Recommended Textbooks:
1. "Digital Signal Processing" by John G. Proakis and Dimitris G. Manolakis
2. "Discrete-Time Signal Processing" by Alan V. Oppenheim and Ronald W. Schafer
3. "Adaptive Filter Theory" by Simon Haykin
4. "Wavelets and Signal Processing" by Lokenath Debnath
Prerequisites:
- Completion of Digital Signal Processing Level One or equivalent
- Familiarity with Fourier transforms, Z-transforms, and basic filter design
- Basic programming skills (preferably MATLAB or Python)
