Quote from: btchip on February 28, 2015, 01:19:18 PM

FPGAs are actually harder to secure than any other kind of hardware - they're designed for flexible & fast hardware emulation, not security (and add proprietary bitstreams on top of that, which make it harder to understand what the chip is actually doing - f.e. see https://eprint.iacr.org/2014/649.pdf).

I fail to see what kind of added security an FPGA would bring, compared to another microcontroller performing the same checks using a totally different code base than the wallet code.

FPGAs are used today in many markets instead of ASICs, such as 10-100 GbEthernet, wireless communication, high throughput encryption, even in smartphones (project ARA). As mentioned, their ability to implement any logic function is a great asset. But after the FPGA is configured at power-on, it is not possible to change it's function over the input-output pins.

However, microcontroller is prone to attacks from it's input-output pins and dedicated interfaces (uart, usb, etc.). If a remote attacker was able to change the code on the microcontroller flash, only a single successful attack is needed. Afterwards the infected microcontroller can leak private keys, attack any device it is connected to (see recent usb vulnerability) and more without the user knowing it. (example: http://www.bunniestudios.com/blog/?p=3554)

With FPGA a developer controls every single bit and logic gate inside the chip as a routine.
With microcontroller you have to trust the compiler and effectively can not check the resulting binary code, because it is very different from the high-level C code written by developer.

FPGA configuration can be done only over dedicated pins of the chip, which are not accessible from external interface.

Moreover, we will publish the source code of the firewall, so that anyone can compile it on our development kit and validate the functionality.

You are welcome to subscribe at our site to receive in-depth design documentation to be published gradually.

www.eliptibox.com