Motorola DSP Developer's Kit

decimate-fir-r.asm

Y = mot###_decimate( X, r, nfilt, 'fir' )

Description

This function decimate resamples data at a lower rate after lowpass FIR filtering. Input vector X is a real vector

Input/Output

Input Parameters: Real vector X, int r, int nfilt

Output Parameters: Real vector Y

Algorithm

The MEX function decimate_fir.m calculates vector b, vector list, int nout, and int nbeg. The CMEX function loads these parameters into dsp memory for asm function use.

nd = length(idata);
m = size(idata,1);
nout = ceil(nd/r);
b = fir1(nfilt,1/r);
gd = grpdelay(b,1,8);
list = round(gd(1)+1.25):r:nd;
lod = length(list);
nlen = nout-lod;
nbeg = r-(nd-list(length(list)));

Assembly function decimate-fir-r.asm then follows these steps to 
calculate decimated vector Y:
nfilt = nfilt+1
itemp = 2*idata(1)-idata((nfilt+1):-1:2)

[odata,zf] = filter(b,1,idata,zi)
odata = odata(list)

Memory & Register

Memory allocation:

• Input vector X is located in X memory
• Input r is loaded in register N3
• Input nfilt is loaded in register N5
• Output vector Y is located in Y memory

In X memory:

• X:(#INA) stores the vector idata
• X:(#INA+(length of idata)) stores the vector b
• X:(#INA+(length of idata)+(length of b)) stores the vector zi/zf
• X:(#INA+(length of idata)+(length of b)+(length of zi)) stores nout
• X:(#INA+(length of idata)+(length of b)+(length of zi)+1) stores the length of list
• X:(#INA+(length of idata)+(length of b)+(length of zi)+2) stores the vector list

In Y memory:

• Y:(#INB) stores the vector odata
• Y:(#INB+(length of idata)) stores the vector itemp

The length of itemp is 2*(nfilt+1)

Input data length:

Length of idata is n

Length of odata is n -> n/r

Length of tb is nfilt+1

Length of zi,zf is nfilt+1

Length of list is = n/r

Length of itemp is 2*(nfilt+1)

Register usage:

• N2 stores the length of idata (n)
• N3 stores r
• N5 stores nfilt
• R2 stores the length of vector b (nfilt+1)

Status Register

No status registers or bits are set explicitly during function execution.

Data Size Limit

Input vector X length must longer than r*(nfilt+1). The length of vector X can't be larger than the continuous available data memory size.

Data Range Limit

The value range of input vector X is [-1.0, +1.0].

Precision

In the case of DSP563, precision is 21 bits.

In the case of DSP566, precision is 12 bits.

conv-c.asm

decimate-fir-c.asm