Other meanings of Aliasing
Signal Processing
In signal processing and related fields, aliasing is an effect that causes different signals to become indistinguishable (or aliases of one another) when sampled. It also refers to the distortion or artifact that results when the signal reconstructed from samples is different from the original continuous signal. Aliasing can occur in signals sampled in time, such as digital audio or video, or in spatially sampled images. The phenomenon is governed by the Nyquist–Shannon sampling theorem, which states that a signal must be sampled at a rate at least twice its highest frequency to be perfectly reconstructed; failure to do so leads to overlapping spectral copies and irretrievable loss of information.1
Aliasing arises from the periodic nature of the spectrum of a sampled signal. When a continuous-time signal is sampled at intervals T, its Fourier transform becomes periodic with period 1/T. If the original signal's spectrum extends beyond the Nyquist frequency (half the sampling rate), the periodic copies overlap, and the overlapping components add together, corrupting the spectrum. The Nyquist–Shannon sampling theorem provides the condition for alias-free sampling: the signal must be bandlimited to frequencies below the Nyquist frequency, and the sampling rate must be at least twice the highest frequency component.1 In practice, signals are rarely perfectly bandlimited, so anti-aliasing filters are used to attenuate high-frequency content before sampling.
Aliasing appears in many domains. In digital audio, insufficient sampling rates cause high-frequency tones to fold down to lower frequencies, producing a characteristic 'warbling' or 'beating' effect. In video and photography, spatial aliasing manifests as moiré patterns when fine periodic textures (e.g., fabric or brick walls) are sampled by a camera sensor. In computer graphics, aliasing appears as jagged edges (jaggies) on rendered objects; anti-aliasing techniques such as supersampling and multisample anti-aliasing are used to smooth these edges.2 In medical imaging, aliasing in MRI can cause ghosting artifacts, and in radar and sonar, it can lead to ambiguous range or velocity measurements.
Beyond the common examples, aliasing has subtle and surprising implications. In the early days of digital audio, some engineers deliberately exploited aliasing to create 'foldback' effects in synthesizers, producing intentionally harsh timbres. In astronomy, aliasing can affect the analysis of time-series data from variable stars, leading to incorrect period determinations. In seismology, spatial aliasing of surface waves can distort tomographic images. A lesser-known historical note: the concept of aliasing predates Shannon—Harry Nyquist's 1924 paper on telegraph transmission already contained the essential idea, and the term 'aliasing' itself was popularized in the 1970s by researchers in digital signal processing.3 In optics, aliasing in lensless imaging systems can be mitigated by computational methods that exploit the known structure of the scene.
Preventing aliasing requires a combination of filtering and oversampling. Anti-aliasing filters are low-pass filters placed before the sampler to remove frequencies above the Nyquist limit. In oversampling converters, the signal is sampled at a rate much higher than the Nyquist rate, then decimated after digital filtering, which relaxes the requirements on the analog filter. In image processing, optical low-pass filters are placed in front of camera sensors to blur fine details that would otherwise alias. In software, resampling algorithms use interpolation kernels with good frequency response to minimize aliasing when changing sample rates. The choice of filter design involves trade-offs between passband ripple, stopband attenuation, and phase linearity.4
Aliasing is a fundamental concept in information theory and signal processing, with wide-ranging implications from audio engineering to medical imaging.
Help improve the encyclopedia. Reports go straight to the site manager.