<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Posts on</title><link>https://blog.amuralmanji.com/posts/</link><description>Recent content in Posts on</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sun, 18 Oct 2026 08:00:00 +0400</lastBuildDate><atom:link href="https://blog.amuralmanji.com/posts/index.xml" rel="self" type="application/rss+xml"/><item><title>Skipjack's S-box and Seed 48571</title><link>https://blog.amuralmanji.com/posts/skipjack_seed_48571/</link><pubDate>Sun, 18 Oct 2026 08:00:00 +0400</pubDate><guid>https://blog.amuralmanji.com/posts/skipjack_seed_48571/</guid><description>&lt;p&gt;One line of &lt;code&gt;srand48&lt;/code&gt; reproduces two-thirds of the NSA cipher&amp;rsquo;s lookup table. Here is what that shows, what it doesn&amp;rsquo;t, and an interactive way to see the difference.&lt;/p&gt;&#10;&lt;blockquote&gt;&#10;&lt;p&gt;&lt;strong&gt;In short.&lt;/strong&gt; Seeding the POSIX &lt;code&gt;drand48&lt;/code&gt; generator with 48571 and running a simple shuffle produces a table that matches &lt;strong&gt;173 of 256&lt;/strong&gt; bytes of Skipjack&amp;rsquo;s F-table. Neighbouring seeds match 1. The remaining 83 bytes can be fixed with 52 swaps, but those swaps are read off the official table, so they would &amp;ldquo;fix&amp;rdquo; any seed. The match is the evidence; the swaps are bookkeeping. One thing the swaps do reveal: they raise nonlinearity from 92 to 100 while differential uniformity gets slightly worse, from 10 to 12.&lt;/p&gt;&#10;&lt;/blockquote&gt;</description></item><item><title>Rediscovered Affine Equivalence Between CLEFIA and Rijndael S-boxes</title><link>https://blog.amuralmanji.com/posts/ae_clefia_aes/</link><pubDate>Wed, 15 Oct 2025 19:30:00 +0400</pubDate><guid>https://blog.amuralmanji.com/posts/ae_clefia_aes/</guid><description>&lt;p&gt;An interesting Link shows that the &lt;strong&gt;CLEFIA S-box&lt;/strong&gt; and the &lt;strong&gt;Rijndael (AES) S-box&lt;/strong&gt; are &lt;strong&gt;affine equivalent&lt;/strong&gt;—meaning one can be transformed into the other through invertible affine transformations over the finite field $\mathbb{F}_2^{8}$.&lt;/p&gt;&#10;&lt;p&gt;Formally, two functions ( f ) and ( g ) are affine equivalent if there exist invertible affine mappings ( A ), ( B ), and constants ( a ), ( b ) such that:&lt;/p&gt;&#10;&lt;p&gt;$$&#10;f(x) = (B \circ g \circ A)(x \oplus a) \oplus b&#10;$$&lt;/p&gt;</description></item><item><title>Understanding Pairwise Independence in Cryptography</title><link>https://blog.amuralmanji.com/posts/pairwise_independent/</link><pubDate>Sun, 07 Sep 2025 16:00:00 +0400</pubDate><guid>https://blog.amuralmanji.com/posts/pairwise_independent/</guid><description>&lt;p&gt;Pairwise independence may look like a simple probability idea, but it plays an important role in proving the security of block ciphers such as AES.&lt;/p&gt;&#10;&lt;hr&gt;&#10;&lt;h2 id="1-pairwise-independence-and-ideal-permutations"&gt;1. Pairwise Independence and Ideal Permutations&lt;/h2&gt;&#10;&lt;p&gt;Two random variables are independent if knowing one does not give information about the other. A set of random variables is &lt;strong&gt;pairwise independent&lt;/strong&gt; if every pair is independent, though this does not mean full independence across all of them.&lt;/p&gt;</description></item><item><title>Implementation of CLEFIA S-box Using Normal Basis — Inspired by A Very Compact Rijndael S-box</title><link>https://blog.amuralmanji.com/posts/clefia_sbox_normal_basis/</link><pubDate>Fri, 18 Apr 2025 14:00:00 +0400</pubDate><guid>https://blog.amuralmanji.com/posts/clefia_sbox_normal_basis/</guid><description>&lt;p&gt;A few years ago, I explored one of my favorite cryptographic building blocks: the substitution box (S-box). Back then, I came across the elegant paper &lt;em&gt;A Very Compact Rijndael S-box&lt;/em&gt;, which showed how to eliminate lookup tables by leveraging &lt;strong&gt;normal basis representation&lt;/strong&gt; and logical operations in $F_{2^{8}}$. That approach deeply resonated with me—compact, efficient, and mathematically sound.&lt;/p&gt;&#10;&lt;p&gt;Recently, I decided to revisit that work—but this time with a new challenge: re-implementing the &lt;strong&gt;CLEFIA S-box&lt;/strong&gt; using the same principles. CLEFIA, Sony’s lightweight block cipher, uses two $8 \times 8$ S-boxes ($S_0$ and $S_1$), and I focused on reconstructing &lt;strong&gt;S₁&lt;/strong&gt; using field inversion and affine transformations—all without tables.&lt;/p&gt;</description></item></channel></rss>