interp-c.asm (Motorola DSP Developer's Kit)

Y = mot###_interp(IDATA, R, LEN, B)

Description

This function resamples data at a higher rate using lowpass interpolation. The input data is complex data.

Y = INTERP (IDATA, R, LEN, B) resamples the sequence in vector IDATA at R times the original sample rate. The resulting resampled vector Y is R times longer, LENGTH(Y) = R*LENGTH(IDATA). A symmetric filter, B, allows the original data to pass through unchanged and interpolates between so that the mean square error between them and their ideal values is minimized.

Input/Output

Input: Complex vector IDATA, int R, int LEN, and vector B

Output: Complex vector Y

Algorithm

int I;
for ( I = 0; I < lengthIDATA*R; I ++)
{
   Y_REAL [I] = 1;
   Y_IMAGE [I] = 1;
}
for ( I = 0; I < lengthIDATA ; I = I ++)
{
   Y_REAL[I*R] = IDATA_REAL[I];
   Y_IMAGE[I*R] = IDATA_IMAGE[I];
}
for( I = 0; I < 2*LEN*R; I ++)
{
   
OD_REAL[I] = 0;
   OD_IMAGE[I] = 0;
}

for (I = 0; I < 2 * LEN; I++)
{
   OD_REAL[I*R] = 2*IDATA_REAL [LEN-1] -IDATA_REAL [2*LEN - I];
   OD_IMAGE[I*R] = 2*IDATA_IMAGE [LEN-1] -IDATA_IMAGE [2*LEN - 
I];
}
/* Call filter  */

[OD, ZI] = filter(B, LEN, OD); 
[Y, ZF] = filter(B, LEN, Y, ZI]; 
for( I = 0, I < (lengthIDATA -LEN) * R; I++)
{
   Y_REAL[I] = Y_REAL[LEN*R+I];
Y_IMAGE[I] = Y_IMAGE[LEN*R+I];
}
for( I = 0; I < 2*LEN*R; I ++)
{
   OD_REAL[I] = 0;
   OD_IMAGE[I] = 0;
}

for (I = 0; I < 2*LEN; I++)
{
OD_REAL[I*R] = 2*IDATA_REAL[lengthIDATA] - IDATA_REAL[lengthIDATA 
-1 -I];
OD_IMAGE[I*R] = 2*IDATA_IMAGE[lengthIDATA] - 
IDATA_IMAGE[lengthIDATA -1 -I];
}
OD = filter (B, LEN, OD, ZF); 
for(I=0;I < LEN*R; I++)
{
   Y_REAL[lengthIDATA*R - LEN*R + I] = OD_REAL[I];
   Y_IMAGE[lengthIDATA*R - LEN*R + I] = OD_IMAGE[I];
}

Memory & Register

Memory allocation:

X memory:

#INA stores the start address of vector idata

#INA+2*n stores the start address of vector tb

Y memory:

#INB stores the start address of vector odata

#INB+2*n*r stores the start address of vector od

N2 stores the location of n

N3 stores the location of r

N5 stores the location of l

R2 stores the length of vector tb (2*r*l+1)

N7 stores the location of n*r

N0, N1, N4, and N6 are used for offset addressing

Status Register

The assembly function interp-c.asm does not set explicitly any status registers/bits during the function execution.

Data Size Limit

The length of vector IDATA can't be larger than the continuous available data memory size.

Data Range Limit

The value of input vectors must be between -1.0 and +1.0.

Precision

In the case of DSP563, precision is 21 bits.

In the case of DSP566, precision is 12 bits.

interp-r.asm log-r.asm