View Raw SPL
/*****************************************************************************
* *
* BLOCKMIN.SPL Copyright (C) 2014 DSP Development Corporation *
* All Rights Reserved *
* *
* Author: Randy Race *
* *
* Synopsis: non-overlapping block minimum *
* *
* Revisions: 13 May 2014 RRR Creation *
* *
*****************************************************************************/
#if @HELP_BLOCKMIN
BLOCKMIN
Purpose: Computes a moving non-overlapping block minimum of a series.
Syntax: BLOCKMIN(s, n)
s - A series, the input data.
n - An integer, the block size to find the minimum.
Returns: A series, the N point moving block minimum.
Example:
W1: 1..12
W2: blockmin(w1, 2)
W2 == {1, 3, 5, 7, 9, 11}
Example:
W3: 1..12
W4: blockmin(w1, 3)
W4 == {1, 4, 7, 10}
Remarks:
BLOCKMIN computes the N point non-overlapping block minimum of
a series.
If the block size is not an integral fraction of the
series length, the end points are not processed.
See XBLOCKMIN to specify the block size as a duration.
See Also:
Blockavg
Blockmax
Decimate
Movmin
Xblockmin
#endif
/* N point block minimum */
blockmin(s, n, flag1 = "", flag2 = "", flag3 = "")
{
local y;
if (argc < 2)
{
error(sprintf("%s - input series and integer blocksize required", __FUNC__));
}
/* decimate the N point moving minimum starting at the Nth sample */
y = movminimum(s, n, n, n, flag1, flag2);
return(y);
}