view libgsmefr/log2.c @ 242:f081a6850fb5

libgsmfrp: new refined implementation The previous implementation exhibited the following defects, which are now fixed: 1) The last received valid SID was cached forever for the purpose of handling future invalid SIDs - we could have received some valid SID ages ago, then lots of speech or NO_DATA, and if we then get an invalid SID, we would resurrect the last valid SID from ancient history - a bad design. In our new design, we handle invalid SID based on the current state, much like BFI. 2) GSM 06.11 spec says clearly that after the second lost SID (received BFI=1 && TAF=1 in CN state) we need to gradually decrease the output level, rather than jump directly to emitting silence frames - we previously failed to implement such logic. 3) Per GSM 06.12 section 5.2, Xmaxc should be the same in all 4 subframes in a SID frame. What should we do if we receive an otherwise valid SID frame with different Xmaxc? Our previous approach would replicate this Xmaxc oddity in every subsequent generated CN frame, which is rather bad. In our new design, the very first CN frame (which can be seen as a transformation of the SID frame itself) retains the original 4 distinct Xmaxc, but all subsequent CN frames are based on the Xmaxc from the last subframe of the most recent SID.
author Mychaela Falconia <falcon@freecalypso.org>
date Tue, 09 May 2023 05:16:31 +0000
parents 68f772469ba7
children
line wrap: on
line source

/*************************************************************************
 *
 *   FUNCTION:   Log2()
 *
 *   PURPOSE:   Computes log2(L_x),  where   L_x is positive.
 *              If L_x is negative or zero, the result is 0.
 *
 *   DESCRIPTION:
 *        The function Log2(L_x) is approximated by a table and linear
 *        interpolation. The following steps are used to compute Log2(L_x)
 *
 *           1- Normalization of L_x.
 *           2- exponent = 30-exponent
 *           3- i = bit25-b31 of L_x;  32<=i<=63  (because of normalization).
 *           4- a = bit10-b24
 *           5- i -=32
 *           6- fraction = table[i]<<16 - (table[i] - table[i+1]) * a * 2
 *
 *************************************************************************/

#include "gsm_efr.h"
#include "typedef.h"
#include "namespace.h"
#include "basic_op.h"
#include "no_count.h"
#include "sig_proc.h"

#include "log2.tab"     /* Table for Log2() */

void Log2 (
    Word32 L_x,         /* (i) : input value                                 */
    Word16 *exponent,   /* (o) : Integer part of Log2.   (range: 0<=val<=30) */
    Word16 *fraction    /* (o) : Fractional part of Log2. (range: 0<=val<1) */
)
{
    Word16 exp, i, a, tmp;
    Word32 L_y;

    test (); 
    if (L_x <= (Word32) 0)
    {
        *exponent = 0;          move16 (); 
        *fraction = 0;          move16 (); 
        return;
    }
    exp = norm_l (L_x);
    L_x = L_shl (L_x, exp);     /* L_x is normalized */

    *exponent = sub (30, exp);  move16 (); 

    L_x = L_shr (L_x, 9);
    i = extract_h (L_x);        /* Extract b25-b31 */
    L_x = L_shr (L_x, 1);
    a = extract_l (L_x);        /* Extract b10-b24 of fraction */
    a = a & (Word16) 0x7fff;    logic16 (); 

    i = sub (i, 32);

    L_y = L_deposit_h (table[i]);       /* table[i] << 16        */
    tmp = sub (table[i], table[i + 1]); /* table[i] - table[i+1] */
    L_y = L_msu (L_y, tmp, a);  /* L_y -= tmp*a*2        */

    *fraction = extract_h (L_y);move16 (); 

    return;
}