References: Windowing, Spectral Leakage, and Peak Detection
-
Window function - Wikipedia - Catalogs the Hann, Hamming, and Blackman windows with their main-lobe-width and side-lobe-level tradeoffs, directly underlying this chapter's window comparison table and window table precomputation.
-
Spectral leakage - Wikipedia - Explains how a signal that does not complete a whole number of cycles within a captured frame spreads its energy across neighboring bins, the exact edge-discontinuity mechanism this chapter opens with.
-
Pitch detection algorithm - Wikipedia - Surveys approaches to estimating a signal's fundamental frequency, including frequency-domain peak-based methods, framing this chapter's argmax-search-plus-parabolic-interpolation pipeline within the broader field of pitch detection.
-
Discrete-Time Signal Processing (3rd Edition) - Alan V. Oppenheim and Ronald W. Schafer - Prentice Hall - This text's widely adopted comparison table of window main-lobe width versus side-lobe level, built on Fredric Harris's landmark 1978 windowing survey, is the standard reference other DSP textbooks' own window tradeoff tables are modeled on.
-
Spectral Audio Signal Processing - Julius O. Smith III - W3K Publishing - Smith is credited with the widely reused closed-form derivation of quadratic (parabolic) interpolation through three adjacent spectral samples for sub-bin peak-frequency estimation, the exact formula this chapter uses for parabolic interpolation.
-
Understanding FFTs and Windowing - National Instruments - Explains how windowing tapers a captured record toward zero at its edges to reduce spectral leakage, and surveys Hann, Hamming, and Blackman-Harris windows and when to choose each, mirroring this chapter's window selection guidance.
-
Window Functions in Spectrum Analyzers - Tektronix - Compares Hann, Hamming, and Blackman-Harris windows by side-lobe suppression and main-lobe width, illustrating the identical no-free-lunch window tradeoff this chapter states as a fundamental, unavoidable relationship.
-
Spectral Analysis of Signals - dspguide.com (Steven W. Smith) - Free chapter covering spectral leakage, the resolution-versus-leakage tradeoff of different windows, and frequency resolution limits, closely paralleling this chapter's treatment of the same topics with additional worked spectral examples.
-
FFT frequency measurement resolution by parabolic and other interpolations - CERN (M. Gasior) - Technical note quantifying how much parabolic interpolation improves frequency-measurement resolution beyond raw bin width, directly supporting this chapter's sub-bin accuracy claims for parabolic interpolation.
-
Windowing - Fundamentals of Signal Processing - Vibration Research University - Explains why an uncorrected edge discontinuity causes leakage and gives practical guidance for choosing between rectangular, flat-top, Hann, Hamming, and Blackman windows, reinforcing this chapter's window-selection decision.