I'm doing this as my final year main project.Does this topic is a good one?

 

so, your plan is to encrypt stuff by sending it through an uart transceiver?

 

You're not off to a very good start due to a woefully inadequate level of detail.

 

so, your plan is to encrypt stuff by sending it through an uart transceiver?

yes, n this des algorithm will be burned on XC3S400

 

Uart is a kind of parallel to serial converter, not an encryption algorithm. Please describe in detail what are you talking about.

 

Actually its base paper is an iee paper on 'design and stimulation of uart serial communication module based on VHDL'.As an application here we do data encryption n decryption

 

Yes, it sounds like a reasonable project. Keep in mind, however, that UART communication is obsolete, as far as computers are concerned. You can still find UART's in embedded systems.

 

If the the idea is to do real-time encryption and decryption on a serial data stream you might want to look at something with more current application like SATA or USB.

 

what is this SATA?

 

Hello,

Have a look here:
http://en.wikipedia.org/wiki/Serial_ATA

Bertus

 

what is this SATA?

A serial interface for transferring data to and from a rotating hard drive, SSD, or Compact Flash chip. I can see numerous interesting applications for such a device especially if you can find a way to make decryption difficult without the proper keys.

In particular you need to devise a method to guarantee there is no backdoor for the NSA or "other" government agency like B-613 to penetrate the device.

 

If you use SATA or USB (or other complex interfaces) then I strongly suggest using an already coded core (check Opencores website) Add your encription or decription core. Don't try to reinvent one of these interfaces.

UART is simple, if you want to try to code one.

 

What is the main focus of your project? A quick scan of the paper you referenced leads me to think it is about creating a UART on a FPGA. So is your project mainly about FPGAs or mainly about DES and encryption?

My guess is that you intended to replicate the work of that paper and then add DES on top and there is nothing wrong with that as a learning exercise however as others have pointed out UART based serial comms are fairly outdated. Industrial control (where such encrypted comms may be required) tend to use Ethernet for data transfer.

 

yes, n this des algorithm will be burned on XC3S400

In your project, what does the DES have to do with the UART?

I'm serious. The heart of systems engineering is usually modularity and problem decomposition. You should have a UART module that doesn't care if the data transmitted over it is encrypted or not and you should have a DES module that doesn't care if the data is coming to/from a UART or not.

If you are going to put them into a single module (not the same as putting a UART module and a DES module together with a wrapper and calling it a DES/UART module), then it had better be because there is a compelling advantage to doing so. Now, in an educational setting, being able to satisfy a contrived specification along the lines of "the project requires it or I will lose points" is a compelling advantage.

 

A serial interface for transferring data to and from a rotating hard drive, SSD, or Compact Flash chip. I can see numerous interesting applications for such a device especially if you can find a way to make decryption difficult without the proper keys.

In particular you need to devise a method to guarantee there is no backdoor for the NSA or "other" government agency like B-613 to penetrate the device.

The forget DES. While I do not believe that the NSA put a backdoor into DES (all indications are that the mysterious tweaks the NSA made to the S-boxes actually strengthened it against differential cryptanalysis which, at the time, was unknown to anyone except the NSA), its key strength, at 56-bits, is just too low to be secure with today's machines.