Table 2 Upper bounds on the number of blocks 
that can be processed in parallel by an s.p.e.d. 
2D-BDCT. 
indicates a direct or a fast implementation of the DCT. 
is equivalent to a 16-bit fixed point implementation. 
and 
are equivalent to a single precision and a double precision floating point implementations, respectively. We have assumed 
.
that can be processed in parallel by an s.p.e.d. 
2D-BDCT. 
indicates a direct or a fast implementation of the DCT. 
is equivalent to a 16-bit fixed point implementation. 
and 
are equivalent to a single precision and a double precision floating point implementations, respectively. We have assumed 
.From: Encrypted Domain DCT Based on Homomorphic Cryptosystems
| Â |
|
|
| |||
|---|---|---|---|---|---|---|
|
|
|
|
|
|
|
4 | 24 | 12 | 12 | 6 | 7 | 3 |
8 | 23 | 8 | 11 | 4 | 7 | 2 |
16 | 22 | 6 | 11 | 3 | 7 | 1 |
32 | 21 | 4 | 11 | 2 | 6 | 1 |
64 | 20 | 4 | 11 | 2 | 6 | 1 |
