Konvolutionen

Konvolutionen, in English convolution, describe a mathematical operation that blends two functions to produce a third that expresses how the shape of one is modified by the other. In both continuous and discrete settings, the operation is central to linear time-invariant systems, signal processing, and statistical modeling.

Continuous convolution of two functions f and g on the real line is defined by (f * g)(t) = ∫_{-∞}^{∞} f(τ) g(t − τ) dτ. Discrete convolution of sequences f[n] and g[n] is (f * g)[n] = Σ_k f[k] g[n − k], with sums taken over the appropriate index range.

Konvolution is commutative, associative, and distributive over addition. An identity element is the delta function δ, since f * δ = δ * f = f. The operation preserves linearity; reversing the order of integration or summation yields the same result given suitable conditions.

The convolution theorem relates convolution to multiplication in the frequency domain: the Fourier transform of a convolution equals the product of the Fourier transforms, F{f * g} = F{f} · F{g}; the inverse transform recovers the convolution. This holds in both one-dimensional and multi-dimensional (e.g., 2D for images) cases.

In practice, convolution is implemented with finite support, requiring boundary handling and padding (zero-padding, reflection). For long signals, FFT-based algorithms reduce complexity from O(NM) to about O(N log N).

Applications include filtering and smoothing of signals and images, blur operations in photography, edge detection with kernels, modeling impulse responses in physics, and, in machine learning, the convolutional layer of neural networks.

The concept originates in Fourier analysis and the study of linear systems; it has become a fundamental tool across mathematics, engineering, and data science.