Multiplier based on operand clipping
An operand and multiplier technology, applied in the field of approximate computing, can solve the problems of tediousness, affecting operation speed, large area overhead and time delay of multiplication operation, and ensuring calculation accuracy, reducing energy consumption, reducing area overhead and time delay. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0079] First, assume that the two decimal operands are:
[0080] A=85, B=170;
[0081] Expressed in binary, if the register bit width:
[0082] no 1 =n 2 = 8;
[0083] but:
[0084] a 0 =a 2 =a 4 =a 6 = 1,a 1 =a 3 =a 5 =a 7 = 0;
[0085] b 0 =b 2 =b 4 =b 6 =0,b 1 =b 3 =b 5 =b 7 = 1;
[0086] based on n 1 , n 2 The value of , setting the number of interception intervals when performing approximate operations on A and B respectively:
[0087] k=2, l=2;
[0088] Second, set the approximation flag:
[0089] A c =or(a 4 ,a 5 ,a 6 ,a 7 )=1,
[0090] B c =or(b 4 ,b 5 ,b 6 ,b 7 )=1;
[0091] Then, based on the value of k, the 1st and 3rd bits of the binary operand A are removed from the lower bits, and the value of the 4th bit is updated at the same time to obtain a simplified multiplication input item A for multiplication p :
[0092] a' 4 =or(and(a 3 ,a 1 ),a 4 )=1,
[0093]
[0094] Similarly, based on the value of l, the 1st and 3rd ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


