Unconditionally secure computers, algorithms and hardware, such as memories, processors, keyboards, flash and hard drives

TL;DR

This paper discusses integrating Kirchhoff-loop-Johnson-(like)-noise ciphers into digital hardware to achieve unconditional security for data communication and storage, resistant to eavesdropping and physical attacks.

Contribution

It proposes a method to embed unconditionally secure cryptographic techniques directly into hardware components like memories and processors.

Findings

Secure key exchange can occur instantly and be renewed randomly.

Physical disassembly does not compromise security due to hidden random addresses.

Breaking the code requires extraordinary resources and access to internal chip structures.

Abstract

In the case of the need of extraordinary security, Kirchhoff-loop-Johnson-(like)-noise ciphers can easily be integrated on existing types of digital chips in order to provide secure data communication between hardware processors, memory chips, hard disks and other units within a computer or other data processor system. The secure key exchange can take place at the very first run and the system can renew the key later at random times with an authenticated fashion to prohibit man-in-the-middle attack. The key can be stored in flash memories within the communicating chip units at hidden random addresses among other random bits that are continuously generated by the secure line but are never actually used. Thus, even if the system is disassembled, and the eavesdropper can have direct access to the communication lines between the units, or even if she is trying to use a man-in-the-middle attack, no information can be extracted. The only way to break the code is to learn the chip structure, to understand the machine code program and to read out the information during running by accessing the proper internal ports of the working chips. However such an attack needs extraordinary resources and even that can be prohibited by a password lockout. The unconditional security of commercial algorithms against piracy can be provided in a similar way.