Skip to content

GaussianFilter

Ehlers DSP filter smoothing ehlers dsp low-pass

Multi-pole Gaussian low-pass filter for reduced lag.

Usage

Use when smooth symmetric price averaging with near-zero phase shift is needed. Works well as a preprocessing step for spectral analysis indicators.

Background

Gaussian filters are the theoretically optimal lowpass filter for minimizing the product of time-domain duration and frequency-domain bandwidth. Ehlers implements them as cascaded pole filters with Gaussian-function-derived coefficients, achieving very smooth output with excellent stopband attenuation.

Parameters

  • period (default: 14): Critical period
  • poles (default: 4): Number of poles (1-4)

Formula

[ \alpha = -\beta + \sqrt{\beta^2 + 2\beta} ] [ \beta = \frac{1 - \cos(2\pi/P)}{2^{1/(2N)} - 1} ]

Source