The Nyquist-Shannon sampling theorem states that a signal of bandwidth B can be perfectly reconstructed from samples taken at rate f_s ≥ 2B. Below this rate, aliasing occurs — high-frequency components "fold back" and appear as lower frequencies. Quantization (bit depth) introduces uniform error bounded by ±LSB/2; SNR ≈ 6.02·N + 1.76 dB for N bits, giving 24 dB for 4-bit and 98 dB for 16-bit audio. The reconstruction uses the Whittaker-Shannon interpolation formula: x(t) = Σ x[n] sinc(f_s·t - n), where sinc is the ideal low-pass filter.