|
NS32FX161-15 Datasheet, PDF (50/102 Pages) National Semiconductor (TI) – Advanced Imaging/Communication Signal Processors | |||
|
◁ |
3 0 Functional Description (Continued)
VARMAC Vector Aligned Real Multiply
and Accumulate
The VARMAC instruction performs a convolution sum of the
X and Y real vectors The previous value of the accumulator
is used and the result is stored in Z 0
Syntax
EXEC VARMAC
15
11 10
0
10000
100 0000 0000
Operation
aligned real X Y
real Z
real acc A
for (n40 n k LENG
n00)
A 4 A 0 X n low * Y n low 0
X n high * Y n high
Z 0 4 (real) A
Note When PARAM CLR is set to ââ1ââ A is cleared to ââ0ââ prior to the first
addition When PARAM SUB is set to ââ1ââ the ââaââ sign is replaced
by a ââbââ sign
VCMAC Vector Complex Multiply and Accumulate
The VCMAC instruction performs a convolution sum of the
X and Y complex vectors The previous value of the accu-
mulator is used and the result is stored in Z 0
Syntax
EXEC VCMAC
15
10000
Operation
11 10
0
100 0111 0101
complex X Y Z
complex acc A
for (n40 n k LENG n00)
A4A0Xn *Yn
Z 0 4 (complex) A
Note When PARAM COJ is set to ââ1ââ X n is multiplexed by the conjugate
of Y n When PARAM CLR is set to ââ1ââ A is cleared to ââ0ââ prior to
the first addition When PARAM SUB is set to ââ1ââ the ââaââ sign is
replaced by a ââbââ sign
VRLATP Vector Real Lattice Propagate
The VRLATP instruction is used for implementing lattice and
inverse lattice filter operations This instruction is used to
update the propagating values of vector Z
Syntax
EXEC VRLATP
15
10000
Operation
11 10
0
100 0010 1100
real X Y Z
real acc A
A 4 (real acc) Z 0
for (n41 n k LENG
n00)
A4A0Xn11 *Yn11
Z n 4 (real) A
A 4 (real acc) Z n
Note When PARAM SUB is set to ââ1ââ the ââaââ sign is replaced by a ââbââ
sign The LENG parameter for this operation must be greater than 1
VCLATP Vector Complex Lattice Propagate
The VCLATP instruction is used for implementing lattice and
inverse lattice filter operations This instruction is used to
update the propagating values of vector Z
Syntax
EXEC VCLATP
15
10000
Operation
11 10
0
100 1110 1000
complex X Y Z
complex acc A
A 4 (complex acc) Z 0
for (n41 n k LENG n00)
A 4 A 0 X n11 * Y n11
Z n 4 (complex) A
Note When PARAM COJ is set to ââ1ââ X n is multiplied by the conjugate of
Y n When PARAM SUB is set to ââ1ââ the ââaââ sign is replaced by a
ââbââ sign The LENG parameter for this operation must be greater
than 1
3 4 5 12 Multiply-and-Add Instructions
VAIMAD Vector Aligned Integer Multiply and Add
The VAIMAD instruction multiplies corresponding elements
of the X and Y integer vectors and adds or subtracts the
result as an integer value to the integer vector Z This re-
sult is placed in the Z output vector
Syntax
EXEC VAIMAD
15
11 10
0
10000
100 0001 0100
50
|
▷ |