Digital Signal Processing: with selected topics: Adaptive Systems, Time-Frequency Analysis, Sparse Signal Processing
This book is a result of author's thirty-three years of experience in teaching and research in signal processing. The book will guide you from a review of continuous-time signals and systems, through the world of digital signal processing, up to some of the most advanced theory and techniques in adaptive systems, time-frequency analysis, and sparse signal processing. It provides simple examples and explanations for each, including the most complex transform, method, algorithm or approach presented in the book. The most sophisticated results in signal processing theory are illustrated on simple numerical examples. The book is written for students learning digital signal processing and for engineers and researchers refreshing their knowledge in this area. The selected topics are intended for advanced courses and for preparing the reader to solve problems in some of the state of art areas in signal processing. The book consists of three parts. After an introductory review part, the basic principles of digital signal processing are presented within Part two of the book. This part starts with Chapter two which deals with basic definitions, transforms, and properties of discrete-time signals. The sampling theorem, providing the essential relation between continuous-time and discrete-time signals, is presented in this chapter as well. Discrete Fourier transform and its applications to signal processing are the topic of the third chapter. Other common discrete transforms, like Cosine, Sine, Walsh-Hadamard, and Haar are also presented in this chapter. The z-transform, as a powerful tool for analysis of discrete-time systems, is the topic of Chapter four. Various methods for transforming a continuous-time system into a corresponding discrete-time system are derived and illustrated in Chapter five. Chapter six is dedicated to the forms of discrete-time system realizations. Basic definitions and properties of random discrete-time signals are given in Chapter six. Systems to process random discrete-time signals are considered in this chapter as well. Chapter six concludes with a short study of quantization effects. The presentation is supported by numerous illustrations and examples. Chapters within Part two are followed by a number of solved and unsolved problems for practice. The theory is explained in a simple way with a necessary mathematical rigor. The book provides simple examples and explanations for each presented transform, method, algorithm or approach. Sophisticated results in signal processing theory are illustrated by simple numerical examples. Part three of the book contains few selected topics in digital signal processing: adaptive discrete-time systems, time-frequency signal analysis, and processing of discrete-time sparse signals. This part could be studied within an advanced course in digital signal processing, following the basic course. Some parts from the selected topics may be included in tailoring a more extensive first course in digital signal processing as well. About the author: Ljubisa Stankovic is a professor at the University of Montenegro, IEEE Fellow for contributions to the Time-Frequency Signal Analysis, a member of the Montenegrin and European Academy of Sciences and Arts. He has been an Associate Editor of several world-leading journals in Signal Processing. Stankovic (with coauthors) won the Best paper award from the European Association for Signal Processing (EURASIP) for 2017 for a paper published in Signal Processing.
Why Read This Book
You will get a compact, example-driven bridge from classical DSP fundamentals to several modern, research-active topics — adaptive systems, time–frequency analysis, and sparse signal processing — explained by an author with decades of teaching and research experience. The book emphasizes intuition and simple numerical examples so you can quickly apply advanced transforms and algorithms to audio, radar, and communications problems.
Who Will Benefit
Graduate and senior undergraduate students, engineers, and researchers in signal processing, communications, audio/speech, or radar who need a single-volume resource linking core DSP with adaptive, time–frequency, and sparse methods.
Level: Intermediate — Prerequisites: Basic calculus and linear algebra, an introductory course in signals and systems (continuous- and discrete-time), and elementary probability; familiarity with MATLAB or Python is helpful but not required.
Key Takeaways
- Analyze and discretize continuous-time signals and systems and relate their properties to DSP implementations.
- Design and implement classical digital filters and FFT-based spectral analysis for practical audio, radar, and communications tasks.
- Apply time–frequency representations (short-time Fourier transform, Wigner distributions, and wavelets) to nonstationary signals.
- Develop and tune adaptive filtering algorithms (LMS, RLS and variants) for noise cancellation, system identification, and channel equalization.
- Formulate and solve sparse signal reconstruction problems using compressed sensing concepts and sparse transforms.
- Interpret statistical signal processing results to estimate spectra and detect signals in noisy environments.
Topics Covered
- 1. Review of Continuous-Time Signals and Systems
- 2. Sampling Theory and Discretization
- 3. Discrete-Time Signals, z-Transform and System Analysis
- 4. Fourier Analysis, DFT and FFT Algorithms
- 5. Digital Filter Design: FIR and IIR Methods
- 6. Spectral Analysis and Windowing Techniques
- 7. Time–Frequency Analysis: STFT, Spectrograms, and Cohen Class
- 8. Wavelet Transforms and Multiresolution Analysis
- 9. Adaptive Systems: LMS, RLS, and Practical Variants
- 10. Statistical Signal Processing and Estimation
- 11. Sparse Signal Processing and Compressed Sensing
- 12. Applications: Audio/Speech, Radar, and Communications Examples
- Appendices: Numerical Examples, MATLAB/Octave Code, Mathematical Background
Languages, Platforms & Tools
How It Compares
Like Oppenheim & Schafer's Discrete-Time Signal Processing it covers core DSP theory, but Stankovic is shorter and places more emphasis on time–frequency methods and sparse processing; compared with Haykin's Adaptive Filter Theory, it gives broader DSP context rather than focusing solely on adaptive filters.
