Mercurial > hg > aim92
comparison man/man1/conv.1 @ 0:5242703e91d3 tip
Initial checkin for AIM92 aimR8.2 (last updated May 1997).
author | tomwalters |
---|---|
date | Fri, 20 May 2011 15:19:45 +0100 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
-1:000000000000 | 0:5242703e91d3 |
---|---|
1 .TH CONV 1 "1 September 1993" | |
2 | |
3 .SH NAME | |
4 conv \- Convolution. | |
5 | |
6 .SH SYNTAX | |
7 conv [options] signal_file impulse_response_file | |
8 | |
9 .SH DESCRIPTION | |
10 The response of a linear filter to an arbitiary input signal is the | |
11 convolution of the signal with the filter's impulse response. | |
12 The `impulse_response_file' contains the impulse response which characterises the | |
13 filter, and the result of the convolution operation is the response of | |
14 that filter to the signal contained in the `signal_file'. | |
15 | |
16 The `signal_file' must consist of binary 16-bit numbers (ie. shorts). | |
17 The `impulse_response_file' must consist of binary floats. | |
18 The output signal is the same length as the input signal and is written on | |
19 the stdout in binary 16-bit numbers. | |
20 | |
21 If the `signal_file' consists of an impulse train, then the output should | |
22 be a repeated impulse response, ie. should be a repeated version of the | |
23 `response_file'. | |
24 | |
25 | |
26 | |
27 .SH OPTIONS | |
28 | |
29 1. length | |
30 | |
31 The amount of the signal file to convolve in time units (s or ms) or in | |
32 samples (no units). Time units are converted to samples using the given | |
33 `samplerate' option. The string "max" is recognised as the remainder of | |
34 input. | |
35 | |
36 2. domain | |
37 | |
38 Select algorithm for convolution. In the time domain ("domain=time") the | |
39 discrete convolution formula is applied directly as a local averaging | |
40 operation on the input signal with weights obtained by time-reversing and | |
41 shifting the impulse response. The impulse response is assumed to be zero | |
42 for all time outside the given file. In the frequency domain | |
43 ("domain=frequency") the FFT of the whole input signal is multiplied by a | |
44 frequency response function (the FFT of the impulse response), and | |
45 the result inverse FFT'd. | |
46 | |
47 3. scale | |
48 | |
49 A scale factor for scaling the output. | |
50 | |
51 | |
52 .SH "SEE ALSO" | |
53 options smooth fft acf | |
54 | |
55 .SH COPYRIGHT | |
56 .LP | |
57 Copyright (c) Applied Psychology Unit, Medical Research Council, 1995 | |
58 .LP | |
59 Permission to use, copy, modify, and distribute this software without fee | |
60 is hereby granted for research purposes, provided that this copyright | |
61 notice appears in all copies and in all supporting documentation, and that | |
62 the software is not redistributed for any fee (except for a nominal | |
63 shipping charge). Anyone wanting to incorporate all or part of this | |
64 software in a commercial product must obtain a license from the Medical | |
65 Research Council. | |
66 .LP | |
67 The MRC makes no representations about the suitability of this | |
68 software for any purpose. It is provided "as is" without express or | |
69 implied warranty. | |
70 .LP | |
71 THE MRC DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING | |
72 ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL | |
73 THE A.P.U. BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES | |
74 OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, | |
75 WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, | |
76 ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS | |
77 SOFTWARE. | |
78 |