Contributions:FIRFilter: Difference between revisions
| (One intermediate revision by the same user not shown) | |||
| Line 17: | Line 17: | ||
==Parameters== | ==Parameters== | ||
===FIRIntegration=== | ===FIRIntegration=== | ||
This parameter may be one of | |||
;0 - none | |||
:no integration, | |||
;1 - mean | |||
:mean value, | |||
;2 - rms | |||
:root mean square, | |||
;3 - max | |||
:maximum value. | |||
For each sample of input, the FIRFilter computes the convolution of past samples with FIR filter coefficients. | For each sample of input, the FIRFilter computes the convolution of past samples with FIR filter coefficients. | ||
When ''FIRIntegration'' is set to "none", this is also the filter's output. | When ''FIRIntegration'' is set to "none", this is also the filter's output. | ||
When ''FIRIntegration'' is set to "mean", "rms", or "max", the respective computation is performed, and | When ''FIRIntegration'' is set to "mean", "rms", or "max", the respective computation is performed on the convolved signal, and its result returned in the output signal's first sample (element, bin). | ||
===FIRCoefficients=== | ===FIRCoefficients=== | ||
A matrix of FIR filter coefficients. Here, rows correspond to channels, defining an individual FIR filter for each channel. When an empty matrix is specified, the FIRFilter returns its input signal unchanged. | A matrix of FIR filter coefficients. Here, rows correspond to channels, defining an individual FIR filter for each channel. When an empty matrix is specified, the FIRFilter returns its input signal unchanged. | ||
Latest revision as of 16:50, 22 November 2008
Synopsis
A finite impulse response (FIR) filter for temporal filtering.
Location
http://www.bci2000.org/svn/trunk/src/contrib/SignalProcessing/FIR
Versioning
Authors
Dennis McFarland, Juergen Mellinger
Source Code Revisions
- Initial development: 31
- Tested under: 2203
- Known to compile under: 2203
- Broken since: --
Parameters
FIRIntegration
This parameter may be one of
- 0 - none
- no integration,
- 1 - mean
- mean value,
- 2 - rms
- root mean square,
- 3 - max
- maximum value.
For each sample of input, the FIRFilter computes the convolution of past samples with FIR filter coefficients. When FIRIntegration is set to "none", this is also the filter's output. When FIRIntegration is set to "mean", "rms", or "max", the respective computation is performed on the convolved signal, and its result returned in the output signal's first sample (element, bin).
FIRCoefficients
A matrix of FIR filter coefficients. Here, rows correspond to channels, defining an individual FIR filter for each channel. When an empty matrix is specified, the FIRFilter returns its input signal unchanged.
States
None.