290 |
This means that `$N \\log k \\ge h$` |
This means that `$N \\log k \\ge h$` |
291 |
|
|
292 |
- this is a nice theoretical result: it *is* possible to sign anything |
- this is a nice theoretical result: it *is* possible to sign anything |
293 |
without trapdoors |
without trapdoors - full feature set of normal (non-one-time) DSs |
294 |
|
|
295 |
- realistic? How much does this need? |
- realistic? How much does this need? |
296 |
|
|
297 |
- Works with `$k=10$`, `$N=16$` for SHA-1. |
- Works with `$k=10$`, `$N=16$` for SHA-1. SIG LENGTH? |
298 |
|
|
299 |
Random or pseudo-random, but probabilistic |
Ordered |
300 |
------------------------------------------ |
------- |
301 |
|
|
302 |
|
- Keep count of number of signatures made |
303 |
|
|
304 |
|
- use bits of count for choosing |
305 |
|
|
306 |
|
- mustn't lose count! |
307 |
|
|
308 |
|
- can't copy key! |
309 |
|
|
310 |
|
Probabilistic limited |
311 |
|
--------------------- |
312 |
|
|
313 |
Shorter signatures |
Shorter signatures |
314 |
|
|
329 |
- birthday paradox again: must not allow the attacker to have |
- birthday paradox again: must not allow the attacker to have |
330 |
2**30 messages being signed |
2**30 messages being signed |
331 |
|
|
|
Ordered |
|
|
------- |
|
|
|
|
|
- Keep count of number of signatures made |
|
|
|
|
|
- use bits of count for choosing |
|
|
|
|
|
- mustn't lose count! |
|
|
|
|
332 |
Applicability to Digital Publishing |
Applicability to Digital Publishing |
333 |
=================================== |
=================================== |
334 |
|
|