Chapter 21 — Further Reading

Annotated pointers for going deeper on orthogonal matrices, rotations, reflections, and the unitary analogue. Page/section numbers are approximate and edition-dependent; use them as a guide rather than a precise locator.

Core textbooks

  • Gilbert Strang, Introduction to Linear Algebra (5th/6th ed.), Chapter 4.4 ("Orthogonal Matrices and Gram–Schmidt"). The closest match to this chapter's spirit — geometry first. Strang's treatment of $Q^{\mathsf{T}}Q = I$, the rotation/reflection dichotomy, and why orthogonal matrices have $\lVert Q\mathbf{x}\rVert = \lVert\mathbf{x}\rVert$ is the canonical undergraduate presentation. His emphasis on $Q$'s role in QR and least squares connects directly back to Chapter 20.
  • Sheldon Axler, Linear Algebra Done Right (3rd/4th ed.), Chapter 7 ("Operators on Inner Product Spaces"). The proof-led, coordinate-free view. Axler develops isometries abstractly (operators that preserve norms) and proves the structure theorems for them, treating the real-orthogonal and complex-unitary cases in a unified inner-product-space framework. Read this if you want the why behind §21.3 at full generality, and as preparation for Chapter 34.
  • Stephen Boyd & Lieven Vandenberghe, Introduction to Applied Linear Algebra (VMLS), Chapter 5 ("Linear independence") and Chapter 10 ("Matrix multiplication"). The applications-first angle. VMLS frames orthonormal vectors and orthogonal matrices around their computational virtues — easy inversion, numerical stability, distance preservation — exactly the practical themes of §21.2.2 and the case studies. Freely and legally available online.

Numerical and computational

  • Lloyd N. Trefethen & David Bau, Numerical Linear Algebra, Lectures 7, 10, and 16. The definitive treatment of Householder reflections and why orthogonal transformations are the bedrock of stable computation (condition number 1). Lecture 10 builds QR from Householder reflectors — the mechanism behind the re-orthonormalization in Case Study 21.1. Graduate-level but exceptionally clear.
  • Gene Golub & Charles Van Loan, Matrix Computations (4th ed.), Chapter 5. The encyclopedic reference for orthogonal/unitary factorizations and their numerical properties. Use as a lookup, not a read-through.
  • numpy / scipy documentation for numpy.linalg.qr, scipy.linalg.qr, scipy.spatial.transform.Rotation, and scipy.stats.ortho_group (which samples random orthogonal matrices). The Rotation class implements exactly the $\mathrm{SO}(3)$ conversions — matrix, Euler angles, quaternion, axis-angle — discussed in §21.7 and Case Study 21.1.

On rotations in 3D and graphics

  • 3Blue1Brown, Essence of Linear Algebra, video on "Three-dimensional linear transformations." A superb visual companion to §21.7; pairs naturally with our recurring visualizer.
  • For quaternions: the 3Blue1Brown / Ben Eater interactive "Visualizing quaternions" explorable explanation is the gentlest visual route into why games and aerospace store rotations as unit quaternions rather than matrices — the cheaper-to-renormalize encoding mentioned in Case Study 21.1.
  • Any standard real-time rendering reference (e.g. Real-Time Rendering by Akenine-Möller et al., the transforms chapter) covers rotation matrices, the rendering pipeline, and re-orthonormalization in practice, extending Chapter 12.

On the unitary / quantum analogue

  • Michael A. Nielsen & Isaac L. Chuang, Quantum Computation and Quantum Information, Chapter 2 ("Introduction to quantum mechanics"). The standard reference establishing that quantum gates are unitary because they must preserve the state vector's norm (total probability) — the rigorous version of §21.10. Section 2.1's linear-algebra review uses exactly our notation for $U^{*}U = I$.
  • For a gentler on-ramp, see the companion volume's treatment of unitary operators in quantum mechanics, which develops the same $U^{*}U = I$ condition in the context of quantum time-evolution, and connects back to the qubit anchor we have been tracking since Chapter 1.
  • John Watrous, The Theory of Quantum Information, Chapter 1 (free online) for the measure-theoretic and operator-algebraic generalization, once you reach Chapter 34's inner-product spaces.
  • The real-rotation half of this chapter is the daily engine behind rotations in games, where orthogonal matrices and quaternions orient every on-screen object without distortion — the subject of Case Study 21.1.
  • The signal-processing thread (§21.11, the unitary DFT and Parseval's theorem) continues immediately in Chapter 22 (Fourier Series) of this book, where orthogonality moves into infinite-dimensional function spaces.

Historical

  • For the history of orthogonal transformations and the people behind them (Householder's 1958 reflector [verify], the nineteenth-century group-theoretic roots), the historical notes in Strang and the biographical asides in MacTutor History of Mathematics (online) are reliable starting points. Treat any single secondary source's dates with mild caution and corroborate before citing.