| |
On Sat, 9 Mar 2002, David Chen wrote:
> Oh well, it was a funny idea though. I might tackle getting a simple
> symmetrical encryption engine put onto the RCX like IDEA or Blowfish
> as a first step some day. The performance/memory issue does keep
> rearing it's ugly head though, may have to think about coding in
> Native RCX Assembler to get the job done... :)
It really isn't that limited when you consider other platforms that run
full PK systems, such as the Dallas Semiconductor iButton. It has a much
more limited processors, more limited RAM, and you write your code in
Java so it has to be interpreted by the 8051-like processor. Granted, it
does have some hardware to speed up the algorithm, but I would bet an
assembly implementation for the H8 would run at least as fast as the iButton.
For key generation, you don't mind it taking a long time (512 bit RSA on
the iButton takes about 45 minutes), since you only do it very infrequently.
You don't want to have the private key ever existing off-board for
security reasons. For encryption, the PK algorithm is only used to
exchange symmetric cipher keys, so it typically encrypts only one packet
per session.
The symmetric ciphers can be implemented on a quite small environment.
The AES winner (Rijndael) could be implemented on a Z80 using 66 bytes of
RAM and 980 bytes of ROM and on a 6805 with 54/1049 bytes, so I am sure
you can implement a decent symmetric cipher along with the PK code in the
H8, and still leave some room left for other things.
References:
G. Keating, _Performance Analysis of AES candidates on 6805 CPU core_,
AES Round 2 public comment, April 15, 1999, available at the URL
http://members.ozemail.com.au/~geoffk/aes-6805/
F. Sano, et al, _Performance Evaluation of AES Finalists on the High-End
Smart Card_, in The Third AES Candidate Conference, printed by NIST.
http://csrc.nist.gov/encryption/aes/round2/conf3/papers/14-fsano.pdf
John A. Tamplin jat@jaet.org
770/436-5387 HOME 4116 Manson Ave
770/431-9459 FAX Smyrna, GA 30082-3723
|
|
Message is in Reply To:
 | | Re: RCX as a PGP engine
|
| (...) Excellent point, Benton Jackson e-mailed me with the following: (...) Good idea using a external sensor to seed some sort of random number generator function. However, the codebase for PGP is pretty darned large and after looking at the source (...) (23 years ago, 9-Mar-02, to lugnet.robotics.rcx.legos)
|
13 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
