Books

Reading and exercise sets organized by textbook. Three shelves: the mathematics and theory books that make up the first three phases of Course 1; the signal-processing and EE books — four of which Course 1 now works directly (Lyons, Kuo, Bendat & Piersol, and Grewal & Andrews, in the Section 11–14 signal-processing phase), with the rest supplying the recommended reading for the bench labs in Course 2; and the image and audio processing books (Gonzalez & Woods, Zölzer) worked in Course 1’s closing phases (Sections 15–20).

Each book lists a Minimum: the bare-minimum chapters to read for the course that uses it. Specific worked problems are filled in here as I complete them (marked ✓ done where a chapter set is finished); the course pages link here rather than listing exercises themselves. Worked proofs are done by hand on paper first, then typeset.

Mathematics & Theory

The eight books behind the theory foundation, ordered by dependency: linear algebra → numerical linear algebra → real analysis → probability → convex optimization → complex analysis → distribution theory and Fourier transforms → information theory (the groundwork under the compression chapters of the image and audio phases).

  • Linear Algebra Done Right — Axler
    • Minimum — read Ch. 1–9. ✓ done: Ch. 1: §1A (all), §1B (all), §1C (1, 3, 4, 19, 20, 23, 24); Ch. 2: §2A (1, 2, 3, 5, 7, 10, 11, 18, 20), §2B (1, 5, 10), §2C (1, 2, 10, 18, 19, 20).
  • Numerical Linear Algebra — Trefethen & Bau
    • Minimum — read Lectures 1–15, 20–23 (norms, QR, least squares, conditioning, floating point, stability, LU). Problems added as worked.
  • Elementary Analysis — Ross
    • Minimum — read Ch. 1–6. ✓ done: Ch. 1: §1 (1.1, 1.6, 1.12), §2 (2.1, 2.8), §3 (3.3, 3.4, 3.5, 3.6), §4 (4.9, 4.14, 4.15, 4.16); Ch. 2: §8 (8.4, 8.5, 8.6, 8.9, 8.10), §9 (9.9, 9.10, 9.11, 9.12), §10 (10.2, 10.5, 10.6, 10.7, 10.8), §11 (11.1, 11.8), §12 (12.1, 12.10, 12.11).
  • Introduction to Probability — Bertsekas & Tsitsiklis
    • Minimum — read Ch. 1–9 (probability models, random variables, expectation, limit theorems, Markov chains, Bayesian & classical inference). Problems added as worked.
  • Convex Optimization — Boyd & Vandenberghe
    • Minimum — read Ch. 2–4, 9 (convex sets & functions, convex problems, unconstrained minimization; skim Ch. 5 duality/KKT). Problems added as worked.
  • Complex Analysis — Bak & Newman
    • Minimum — read Ch. 1–7 (complex numbers, analytic functions, line integrals & Cauchy’s theorem, entire functions & the Cauchy integral formula, power series, Laurent series and residues). Problems added as worked.
  • A Guide to Distribution Theory and Fourier Transforms — Strichartz
    • Minimum — read Ch. 1–5 (distributions, the Dirac delta, distributional derivatives, convolution, the Fourier transform of tempered distributions). Problems added as worked.
  • Elements of Information Theory — Cover & Thomas
    • Minimum — read Ch. 2 (entropy, relative entropy, and mutual information); the information-theoretic groundwork under the entropy coding of image compression (Course 1 Section 17) and the psychoacoustic audio coding of Section 19, and the cross-entropy/KL training objectives of Course 2’s on-device models. Problems added as worked.

Signal Processing & EE

The signal-processing shelf. Four of these — Lyons, Kuo, Bendat & Piersol, and Grewal & Andrews — are worked as coursework in Course 1 Sections 11–14 (digital and statistical signal processing); they were chosen because each chapter carries a problem set. The rest are the recommended reading behind the bench labs of Course 2: Oppenheim’s Signals and Systems for the continuous-time transform theory, and Hayes for the statistical modeling.

  • Signals and Systems — Oppenheim, Willsky & Nawab
    • Minimum — read Ch. 1–4 (signals, LTI systems, convolution, Fourier series & transform), Ch. 6 (time/frequency characterization, Bode), Ch. 7 (sampling theorem, aliasing, reconstruction). Notes added as worked.
  • Understanding Digital Signal Processing — Lyons
    • Minimum — read Ch. 1–6 (discrete sequences & systems, periodic sampling and aliasing, the DFT, the FFT, FIR filters, IIR filters), Ch. 10 (sample-rate conversion — decimation, interpolation, polyphase, CIC), Ch. 11 (signal averaging — coherent/incoherent averaging and SNR gain), Ch. 12 (digital data formats and their effects — fixed-point binary formats, precision, dynamic range, finite word length), Ch. 13 (DSP tricks — fast FIR filtering, A/D converter testing, spectral peak location). Worked in Course 1 Sections 11–13, as the intuition-first companion to Kuo. Problems added as worked.
  • Statistical Digital Signal Processing and Modeling — Hayes
    • Minimum — read Ch. 3–5 (discrete random processes, signal modeling, the Levinson recursion), Ch. 7–9 (Wiener filtering, adaptive filters, spectrum estimation). Notes added as worked.
  • Real-Time Digital Signal Processing — Kuo, Lee & Tian
    • Minimum — read Ch. 1–7 (real-time DSP systems & the analog interface; sampling, quantization, fixed-point representation & overflow; FIR design and implementation; IIR design and realization; frequency analysis, the DFT & FFT; adaptive filtering and the LMS algorithm; signal generation and detection). Worked in Course 1 Sections 11–12. Problems added as worked.
  • Kalman Filtering: Theory and Practice with MATLAB — Grewal & Andrews
    • Minimum — read Ch. 1–5, 7 (linear dynamic systems & observability; probability, expectancy and the LMSE; random processes and covariance propagation; the linear optimal/Kalman filter, the Kalman–Bucy form and the matrix Riccati equation; and the numerically stable square-root/UD implementations). The recursive, real-time counterpart to Hayes’s Wiener filter. Worked in Course 1 Section 14. Problems added as worked.
  • Random Data: Analysis and Measurement Procedures — Bendat & Piersol
    • Minimum — read the chapters on random-data classification & probability/statistical fundamentals, correlation & power-spectral-density functions, single-input/output frequency-response & coherence, the statistical (random & bias) errors in spectral estimates, and the data-acquisition / spectral-analysis procedures (≈ Ch. 1, 3–6, 8, 11 in the 4th ed.). The measurement-error companion to Hayes. Worked in Course 1 Section 13. Problems added as worked.

Image & Audio Processing

The two applied books worked in Course 1’s closing phases (Sections 15–20): digital image processing in two dimensions and digital audio signal processing in one, each closing on a deep-convolutional-network chapter. Both were chosen because every chapter carries a problem set that can be worked by hand.

  • Digital Image Processing — Gonzalez & Woods
    • Minimum — read Ch. 2–5 (digital image fundamentals; intensity transformations and spatial filtering; filtering in the frequency domain — the 2-D DFT; image restoration and the Wiener filter), Ch. 8 (image compression and entropy coding), Ch. 10–11 (image segmentation — edges, thresholding, clustering, watersheds; feature extraction — descriptors, PCA, SIFT), and Ch. 12 (image pattern classification — neural networks, backpropagation, and deep convolutional neural networks). Ch. 6 (color), Ch. 7 (wavelet and other transforms), and Ch. 9 (morphology) as needed. Worked in Course 1 Sections 15–17 (and Ch. 12 in Section 20). Problems added as worked.
  • Digital Audio Signal Processing — Zölzer
    • Minimum — read Ch. 1–4 (signals and transforms; quantization, dither, and noise shaping; sampling rate conversion; AD/DA conversion and delta-sigma modulation), Ch. 6–10 (equalizers and fast convolution; room simulation and reverberation; dynamic range control; audio coding and psychoacoustics; nonlinear/virtual-analog processing), and Ch. 11 (machine learning for audio — feedforward and convolutional neural networks). Ch. 5 (audio processing systems and interfaces) as reference. Worked in Course 1 Sections 18–19 (and Ch. 11 in Section 20). Problems added as worked.