欢迎访问ic37.com |
会员登录 免费注册
发布采购

PDSP16116BB0AC 参数 Datasheet PDF下载

PDSP16116BB0AC图片预览
型号: PDSP16116BB0AC
PDF下载: 下载PDF文件 查看货源
内容描述: 16 x 16位乘法器复 [16 X 16 Bit Complex Multiplier]
分类和应用:
文件页数/大小: 17 页 / 272 K
品牌: MITEL [ MITEL NETWORKS CORPORATION ]
 浏览型号PDSP16116BB0AC的Datasheet PDF文件第2页浏览型号PDSP16116BB0AC的Datasheet PDF文件第3页浏览型号PDSP16116BB0AC的Datasheet PDF文件第4页浏览型号PDSP16116BB0AC的Datasheet PDF文件第5页浏览型号PDSP16116BB0AC的Datasheet PDF文件第7页浏览型号PDSP16116BB0AC的Datasheet PDF文件第8页浏览型号PDSP16116BB0AC的Datasheet PDF文件第9页浏览型号PDSP16116BB0AC的Datasheet PDF文件第10页  
PDSP16116  
NORMAL MODE OPERATION  
of CLK. The X and Y port registers are individually enabled by  
CEX CEY  
quired to be permanently enabled, then these signals may be  
tied to ground.  
When the MBFP mode select input is held low the ‘Normal’  
mode of operation is selected. This mode supports all complex  
multiply operations that do not require block floating point  
arithmetic.  
the  
and  
signals respectively. If the registers are re-  
Complex two’s complement fractional data is loaded into the  
X and Y input registers via the X and Y Ports on the rising edge  
The Real and Imaginary components of the fractional data  
are each assumed to have the following format:  
Bit Number  
Weighting  
15 14 13 12 11 10  
9
8
7
6
5
4
3
2
1
0
1
2
3
4
5
6
7
8
9
15  
S
222222222210 211 212 213 214 2–  
Where S = sign bit, which has an effective weighting of 220  
The value of the 16-bit two’s complement word is (213S)1(bit143221)1(bit133222)1(bit123223) …  
Multiplier Stage  
Oneachclockcyclethecontentsoftheinputregistersarepassed  
to the four multipliers to start a new complex multiply operation.  
Each complex multiply operation requires four partial products  
(XR3YR), (XR3YI), (XI3YR), (XI3YI), all of which are calculated  
in parallel by the four 16316 multipliers. Only one clock cycle is  
required to complete the multiply stage before the multiplier results  
are loaded into the multiplier output registers for passing on to the  
adder/ subtractors in the next cycle. Each multiplier produces a 31-  
bit result with the duplicate sign bit eliminated. The format of the  
output data from the multipliers is:  
Bit Number  
Weighting  
30 29 28 27 26 25 24  
7
6
5
4
3
2
1
0
1
2
3
4
5
6
30  
S
222222–  
223 224 225 226 227 228 229 2–  
The effective weighting of the sign bit is 220  
Adder/Subtractor Stage  
The 31-bit real and imaginary results from the multipliers  
are passed to two 32-bit adder/subtractors. The adder calcu-  
lates the imaginary result [(XR 3 YI) 1 (XI 3 YR)] and the  
subtractor calculates the real result (XR 3 YR) = (XI 3 YI).  
Each adder/subtractor produces a 32-bit result with the  
following format:  
Bit Number  
Weighting  
31 30 29 28 27 26  
8
7
6
5
4
3
2
1
0
1
2
3
4
30  
S
20 2222–  
222 223 224 225 226 227 228 229 2–  
The effective weighting of the sign bit is 221  
Rounding  
significant 16 bits are unaffected). Inserting a ‘1’ ensures that  
the rounding error is never greater than 1 LSB and that no DC  
bias is introduced as a result of the rounding processes. The  
format of the rounded result is:  
The ROUND control when asserted rounds the most  
significant 16 bits of the full 32-bit result from the shifter. If the  
ROUND signal is active (high), then bit 16 is set to ‘1’, rounding  
the most significant 16 bits of the shifted result. (The least  
Bit Number  
Weighting  
31 30 29 28 27  
18 17 16 15 14 13  
2
1
0
1
2
3
1
17  
30  
S
20 222–  
22 213 214 215 216 2–  
228 229 2–  
ROUNDED VALUE  
LSBs  
The effective weighting of the sign bit is 221  
Result Correction  
Complex Conjugation  
Due to the nature of the fraction two’s complement repre-  
sentation it is possible to represent 21 exactly but not 11. With  
conventional multipliers this causes a problem when 21 is mul-  
tiplied by 21 as the multiplier produces an incorrect result. The  
PDSP16116 includes a trap to ensure that the most positive  
number (value = 1·2230, hex = 7FFFFFFFF) is substituted for  
the incorrect result. The multiplier result is therefore always a  
correct fractional value. Fig.2 shows the value ‘1’ being multi-  
plexed into the data path controlled by four comparators.  
Either the X or Y input data may be complex conjugated by  
asserting the CONX or CONY signals respectively. Asserting  
either of these signals has the effect of inverting (multiplying  
by 21 ) the imaginary component of the respective input. Table 3  
shows the effect of CONX and CONY on the X and Y inputs.  
Operation  
CONX  
CONY  
Function  
Low  
High  
Low  
High  
Low  
Low  
High  
High  
X 3 Y  
(XR 1 XI)3(YR 1 YI)  
Conj. X 3 Y (XR 2 XI)3(YR 1 YI)  
X 3 Conj. Y (XR 1 XI)3(YR 2 YI)  
Invalid  
Invalid  
Table 3 Conjugate functions  
6