FAQ  •   Login  •   Register  •   Subscribe 

Welcome to the Forum for InventorSpot.com, the most popular invention related website in the world. Read our welcome message.

Skip to content

Moderators: Michelle, Scrupulous, citizen


Better than RSA Encryption

Postby Codex » Fri Dec 05, 2008 8:08 pm

Codex
White Belt
 
Posts: 16
Joined: Sun Oct 19, 2008 2:59 am
I have recently finished work on a new algorithm that is more secure than RSA. Impossible, I know , yet there it is. As computer science advances RSA becomes more and more vulnerable. If we do not have a viable option to RSA and ECC cryptography before Quantum Computing becomes viable all of our data will be at risk. A quantum computer could quickly break either encryption. My alternative, MSE, will be able to provide protection against the quantum computer. My question is , where do I go from here? I've been told you simply publish the algorithm after I have filed my P patent. Is that the path that you would recommend? Unlike RSA my algorithm is software dependent (You would not be abet to compute the calculations otherwise)and the code is still being written. The methodology is all worked out however. I suppose I will pitch to RSA Secuity..any ideas?

Postby Levi Porter » Sat Dec 06, 2008 1:26 am

User avatar
Levi Porter
Brown Belt
 
Posts: 889
Joined: Mon Oct 15, 2007 2:37 am
Hi Codex,

Cool 8)

Does the MSE you mentioned = mean squared error? I am somewhat familiar with cryptography. I believe functional prototype quantum computers can only accomplish the mathematical problem of 3 x 5 = 15, and are the size of large desk, so their threat is a ways out.

Have you filed for a patent? I believe you don't have to include the source code for software programs in patent applications. You'd want to verify that.

As far as publishing immediately after filing goes, I don't know that this the best thing to do. If your patent was rejected, you'd be left with an obvious work equaling open source, while the world would have been looking at what you had with some people possibly working on variations of your idea and filing their own patent applications. In the event your app is rejected, and their patent gets awarded, you'd be stuck.

If you really have something that would blow the doors off of RSA and ECC, then I would keep it as quiet as possible. You would have a hot ticket that many people would want. You would have several interested parties.

If you haven't filed yet, I'd make sure to use the Non Publication Request. That way, it only gets published if a patent is issued.

When you say you'd pitch it to RSA, who did you have in mind?

I read somewhere that a US Government agency spent 25 million dollars to buy a bunch of licensing agreements from this Canadian company that has like 350 ECC patents I believe..... http://www.certicom.com/

Am I correct in thinking that the most vulnerable part of encryption is when it is in the air, or is it equally vulnerable when ran through wires or fiber optics?

Postby Codex » Sat Dec 06, 2008 2:37 am

Codex
White Belt
 
Posts: 16
Joined: Sun Oct 19, 2008 2:59 am
First let me thank you for your quality response. :) Actually, my answer to RSA does not involve Mean Squared Error, but unfortunately I can't say what the MSE stands for, as I'm sure you understand. As for quantum computing, Moore's Law says that we should not understate the human desire to further advance our technology. It took only 30 years for computers to shrink from the size of rooms to the desktop.

Thanks for the sage advice about holding off on publication I will follow it. I will also request that the patent request be non disclosed should my application to denied.

The system is only vulnerable as the keys are transmitted but I think all wireless data transmission is more vulnerable than through wire. Further encryption of the keys themselves is also possible with good old RSA or ECC or even as a double layer through MSE.

It has none of the weaknesses of RSA such as timing or branch prediction attacks.
Thanks for the tip about not having to have the code written yet, that is a bit daunting as this is not my full time occupation, I like to code but it's so time consuming. I considered getting a partner but then who can I trust? Mixing business amongst friends is nearly as bad as with strangers. ;)

Postby Codex » Sat Dec 06, 2008 2:44 am

Codex
White Belt
 
Posts: 16
Joined: Sun Oct 19, 2008 2:59 am
Wow and thanks for the link and encouragement.

Postby Levi Porter » Sat Dec 06, 2008 3:46 am

User avatar
Levi Porter
Brown Belt
 
Posts: 889
Joined: Mon Oct 15, 2007 2:37 am
No problem, and thank you for saying what you can. :wink:

I don't know if you have ever filed for a patent before, but, I'm pretty sure the Non Publication Request has to filed when the application is filed with a non provisional.

Also, make sure to verify what I said about the source code.

I think I can figure out some pretty cool encryption. I would love to say what my invention is, but I need to keep it under wraps for now.

Good point about mixing business amongst friends or strangers.

At some point, I am sure you know, you'll have to trust somebody.

Right now, I am pursuing the Government.

Postby Codex » Sat Dec 06, 2008 1:24 pm

Codex
White Belt
 
Posts: 16
Joined: Sun Oct 19, 2008 2:59 am
Actually I haven't. I'm a FOREX broker by trade, I know, every one and their uncle is a 'day trader' these days but this is actually what I do. I became interested in encryption when I saw the RSA challenge numbers and set myself the goal of figuring a way to break it without the use of computers. Obviously I failed but it was a fascinating thought experiment.
My current invention has nothing to do with current encryption really but RSA is what got me interested.
It would be nice if everyone could just exchange their ideas with no consequences but alas, we cannot.

I will verify that bit about the source code, thanks. What I need to get protection for is the novility (If that is a word) of the concept I think. The clock is ticking :(

Good luck with yours!

Postby Levi Porter » Mon Dec 08, 2008 5:40 am

User avatar
Levi Porter
Brown Belt
 
Posts: 889
Joined: Mon Oct 15, 2007 2:37 am
Thank you Codex,

Um, if you are in the US, and can prove the dates of your work, that will qualify for priority that you were the first to create it. Journals and witness signatures can also help. If you performed most of the work on a computer, I think it would be possible to have the hard drive analyzed to verify when and what key strokes were used.

I would think that one of the best ways to protect your novelty is a Provisional Patent. It's only 110.00, and good for 12 months. You would want to include an enabling description so that anyone familiar in the art would know how to build and use your product. An unambiguous claim can help with International Patents I believe. You don't have to include any diagrams in a Provisional, but if you have them, and they help explain it, you might want to include them.

As soon as your Provisional application is received at the USPTO, you have the legal right to say that it is patent pending. It is important to note that you have 1 year from that date to file the full Non Provisional Utility Patent Application, or you loose your right of priority, and cannot refer to it in the full Non Provisional. If 1 year is not enough time for you to get prepared for the Non Provisional, then you might want to wait until you know you can be totally prepared within a 12 month period.

Isn't there a virtually limitless possible number of solutions with cryptography? I am surprised that no one has been able to create a crack proof encryption yet.

If you end up dealing with some government agencies and industry, you'll likely have to get good at describing what you have without divulging your proprietary information.

For me this is really difficult, as explaining what I have is hard enough, let alone explaining it without compromising my rights. This seems to be an art form where you want to say enough to perk interest without cutting your own throat. I feel like saying anything about what I have gives too much of it away.

I'll have to ask around here to see what some good methods might be.

Peace,

Levi

cron