Figure B-5 Lms Adaptive Filter-Double Precision Delayed Memory Map - Motorola DSP56800 Manual

move
_coefupdate
move #-2,n
move
;
;
B.1.7.3 Double Precision Delayed
Figure B-5 shows a memory map for this implementation of the double-precision delayed LMS adaptive
filter.
Figure B-5. LMS Adaptive Filter—Double Precision Delayed Memory Map
; Delayed LMS algorithm with matched coefficient and data vectors
; Algorithm runs in 5N (2 coeffs processed in each 10 cycle loop)
; Data Sample is stored in Y0 and Y1.
; Coefficient is stored in X0
; Loop Gain * Error is stored in X:(R2) (will be placed in X0).
; FIR operation done in B.
; Coeff update operation done in A.
; FIR sum = a = a +c(k)
; c(k) = b = c(k)
new
opt cc
move #state,r0
move #ntaps,m0
move #c,r3
move #c-2,r1
move #0,n
clr b
move
x:(r3)+,a0
x:(r0)+n,y0
Total:
r0
r1,r3
*x(n-k)
old
-mu*e *x(n-k-1)
old old
x:(r0)+,y0
x:(r0)+,y1 x:(r3)+,x0 ; 1
DSP Benchmarks
; 1 1
; 1 1
; 1 1
_______
21 6N+18
X memory
x(n)
x(n-1)
.
x(n-N+1)
c0h
col
c1h
c1l
; 2 2
; 2 2
; 2 2
; 2 2
; 1 1
; 1 1
1
; fetch next c()l
; adjustment for
; filtering
; update r0
emulate (Rn) adr
mode
y0 = x(n)
y1= x(n-1), x0=c0h
AA0083
B-11