I have heard that public key encryption and related cryptologies need
32bit processors to execute at a reasonable rate. If this is not true, please correct me.Could someone recommend a cheap 32bit development board, preferably with an ethernet MAC (I should be able to get transceiver, magnetics, and connector if need be).
Also what are the pros/cons of using a MCU with and without a MMU? Is programming significantly more difficult without one?
--David
That all depends on what you mean by reasonable. An 8-bit uP with a hardware multiply will do quite well. For RSA you need to multiply and take remainders for numbers with more that 1000 bits. That can be done 8 bits at a time.
-Robert Scott Ypsilanti, Michigan (Reply through this forum, not by direct e-mail to me, as automatic reply address is fake.)
Well I need to be able to encrypt 256 bits in less than 5 sec. or generate a 256 bit key with DH key exchange in the same time. Also any suggestions on cheap development kits
--David
In article , David writes
As most smart cards used to be 8 bit (8051 a lot of them) and had RSA you clearly don't need a 32 bit processor. Though to be fair a lot of them also tended to have a crypto engine on them.
BTW define "reasonable" :-)
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ \/\/\/\/\ Chris Hills Staffs England /\/\/\/\/\ /\/\/ snipped-for-privacy@phaedsys.org
"David" wrote
Define 'reasonable'. PKE was common on 4.77MHz 8088 XT's.
The obverse. For a stand-alone application it is easier without one.
-- Nicholas O. Lindan, Cleveland, Ohio Consulting Engineer: Electronics; Informatics; Photonics. Remove spaces etc. to reply: n o lindan at net com dot com psst.. want to buy an f-stop timer? nolindan.com/da/fstop/
"David" wrote
The 32 byte memory requirement shouldn't be much of a strain. A 32-bit processor with an MMU may not be needed.
As for speed, I don't think anyone makes a processor this slow anymore.
Pick anything you want. If this is a 'one-of' then a "Basic Stamp" would work fine. Otherwise, use whatever you have at hand and are familiar with.
-- Nicholas O. Lindan, Cleveland, Ohio Consulting Engineer: Electronics; Informatics; Photonics. Remove spaces etc. to reply: n o lindan at net com dot com psst.. want to buy an f-stop timer? nolindan.com/da/fstop/
Well the problem is: the ones I've seen range from more than my car now ($2000) to more than my car when I bought it a few years ago ($5000).
Is it possible to build yourself a development kit (with a good chance of success) cheaply from an available MCU? Or buy a reasonable priced kit?
As I said, correct me if an 8 or 16 bit MCU will do. Suggestions of suitable kits would be appreciated.
Also I would prefer not having to write my own TCP/IP stack or buying one for $1000+. So if it is a product with a freely available TCP/IP stack, the happier I be. Also what are your thoughts on the MMU?
--David
The device will need to do the following things:
NTP client to update RTC Custom server protocol for configuring device with encryption of communication Main program that will output particular data to certain peripherals (IR/RF encoders -> IR/RF blasters) at particular times as configured by client over TCP/IP through integrated ethernet.
If all this can be smushed into one program then I guess I don't need a MMU or OS. But I don't if it will work well.
--David
A better aproach which I took when implementing a design which used AES as the crypto system was to design an fpga which did the bulk of the encryption/decryption and a smal cheapo processor to get the results on and off the board and handle the key scheduling. This saved a lot of processor time as its very easy to pipeline the aes algo within the fpga.
I implemented a *very* non-optimal "big number" library and DH key exchange, and was amazed at how long it took to execute on an MPC860.
I also implemented DES, which didn't seem too heavy weight. The key exchange was by far the most processor intensive part of the process.
I did this for "fun", and I really don't have any encryption experience other than that. Just thought it might be an "interesting" data point.
-- Michael N. Moran (h) 770 516 7918 5009 Old Field Ct. (c) 678 521 5460 Kennesaw, GA, USA 30144 http://mnmoran.org "So often times it happens, that we live our lives in chains and we never even know we have the key." The Eagles, "Already Gone" The Beatles were wrong: 1 & 1 & 1 is 1
You might have a look at
Regards
-Bill Knight
again, what is reasonable ? Some implementations use FPGAs to do as much in parallel as possible. AFAIK, the Rhyndael/AES is available as core.
Rene
-- Ing.Buero R.Tschaggelar - http://www.ibrtses.com & commercial newsgroups - http://www.talkto.net
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.