#crystals-kyber

[thequantumspaceorg]

EPISODE 13 – Crypto Under Quantum Siege

Please use the player below to hear our latest Innovating Trust podcast or choose your preferred podcast platform below. Crypto Under Quantum Siege Quantum computing is moving from theory to threat and the first systems it will challenge sit at the heart of blockchain security. In this episode of The Quantum Space Podcast – Innovating Trust, we unpack how algorithms such as Shor’s and Grover’s…

EPISODE 13 – Crypto Under Quantum Siege

[businesscraftse-blog]

Cybersecurity in the Quantum Age: Post-Quantum Crypto & Q-Day Preparation

Every era has its defining cybersecurity challenge. In the early 2000s, it was worms and viruses. The 2010s brought ransomware and state-sponsored intrusions. Now, the 2020s have begun a race against time toward what experts ominously call Q-Day—the moment when a sufficiently powerful quantum computer could break widely used public-key cryptography. This is not science fiction. Algorithms like…

Cybersecurity in the Quantum Age: Post-Quantum Crypto & Q-Day Preparation

[astelor]

Fixing my skill issues 2025.05.22

Today I realized I am a bit (very) cooked in my hardware implementation project. I am a little bit new to HDL programming and designing and I don't have much time left before the master's degree application season starts (in about a hundred days). So I down-scoped the project to only IND-CPA encryption plus choosing the ML-KEM-768 security strength, which is implementing essentially all the required hardware modules with the least variations.

Now here are the problems:

It definitely needs a hash function, the extendable one in the SHA3 family

If I don't optimize the area size, there is a chance it won't fit onto the FPGA board. (but it's a non problem since I should aim for correctness first on simulation)

Again, I only have about 100 days from now (I already understand most of the modules except the Hash module)

I get why the TAs are pushing me for hardware designs recently after putting this together in Microsoft Visio.

So here are the things I need to crank out:

Hash function module especially SHAKE256

Modular Reduction module

InvNTT module

Rejection sampling module (for generating the public key matrix A in NTT form)

And I think I will go for implementing SHAKE256 (writing the HDL) first. And I don't think module hardening is possible since I stalled too long for reading the papers :')

[astelor]

Fixing my skill issues 2025.05.21

Today or rather yesterday before I decided to call it a day, I figured out what centered binomial distribution is. And what it mean by sampling an uniform string of bits (essentially a random number)

η (Eta) decides the range of noise array output, which will be [-η, η] and the algorithm crawls through uniform bits.

The generated noise is a key component for Learning-With-Error problem. Especially the key generation and encryption algorithms.

The table of parameter shows that in different strength of security, the η differs slightly. I have not been looking into the software implementation as I am still busy making sense of hardware module designs. Hopefully I finish staring at the hardware designs by the end of May (and oh my there are many things to be done...)

[astelor]

Fixing my skill issues 2025.05.20

It has been weeks since I started studying the CRYSTALS-Kyber key encapsulation scheme. Honestly, the more I look at it the more I realize I might have just scraped the surface. Every time I thought I understand it enough and moved on to the next step (math equations to software implementation to hardware implementation), I realized I did not understand it that well and went back to the specifications.

Good news: I have a better understanding on the hardware schemes.

Bad news: An implementation that is heavily optimized in time might not be possible due to resource constraints (FPGA board).

I might have neglected to mention that I am implementing the scheme on FPGA, and I think I will have to go for encryption module first. Since the resource will have to be reused, I guess finishing an encapsulation module is completing half of the decapsulation module.

Before going back to the specs and pseudo-code, I forgot the IND-CPA encryption scheme requires hash functions to generate the noise vectors. It is not a cryptographic implementation if I don't include hash functions.

The progress is unfortunately slow because of my skill issues and the many things I need to read.