High Performance Significance Approximation Error Tolerance Adder for Image Processing Applications
Citations
22 citations
17 citations
15 citations
12 citations
10 citations
References
1,361 citations
"High Performance Significance Appro..." refers methods in this paper
...The approximation techniques were developed specifically for the digital image processing that can be found in parallel (or block) processing to either increase the effective throughput or to reduce the power consumption of the original image processing systems [7]....
[...]
386 citations
"High Performance Significance Appro..." refers background or methods in this paper
...In the approximation logic, approximation is chosen since it gives minimum errors when compared to other approximations [1]....
[...]
...total error generated can be computed by the following equation [1]:...
[...]
...A low performance design might accept the potential errors, or be tolerant of them to some threshold; a high performance design can use the approximate result to speculatively execute instead of ideally waiting for the exact result to become available [1]....
[...]
...The carry-select adder is simple but rather fast, having a gate level depths of adding two n-bit numbers with a carry select adder is done with two adders (two ripple carry adders) in order to perform the calculation twice, one time with the assumption of the carry being zero and the other assuming one [1]....
[...]
286 citations
"High Performance Significance Appro..." refers background in this paper
...It takes in three inputs and creates two outputs, a sum and a carryout [6, 8, 11, 14]....
[...]
211 citations
"High Performance Significance Appro..." refers background in this paper
...Some applications need an approximation adder to operate at the high frequencies such as video processing, whereas some other applications require high throughput and accuracy with a low-power circuit such as multiple-input and multipleoutput systems [3, 10]....
[...]
196 citations
"High Performance Significance Appro..." refers background in this paper
...It takes in three inputs and creates two outputs, a sum and a carryout [6, 8, 11, 14]....
[...]