Encoding and Decoding Messages Using Matrices

Name: [Your Name]

Class: [Class/Grade]
Date: [Date]

Project Writeup

Cryptology and Matrices

Cryptology is the branch of mathematics and computer science dealing with secure communication
(encryption) and code-breaking (cryptanalysis). In general, it is defined as “the practice and study of
techniques for secure communication in the presence of adversarial behavior” 1 . Cryptology uses math
(especially number theory and algebra) to design ciphers. As one source explains, “cryptology is the
mathematics, such as number theory and the application of formulas and algorithms, that underpin
cryptography and cryptanalysis” 2 . In a cipher, readable plaintext is converted into ciphertext using a
secret key; only someone who knows the key (and the decryption method) can recover the original
message. Modern cryptography is heavily based on mathematics 3 , and many ciphers use algebraic
structures like matrices.

A classic example of matrix-based encryption is the Hill cipher (invented by Lester Hill, 1929). In a Hill
cipher, the plaintext is broken into blocks of fixed size (say n letters). Each block is converted to a numeric
vector, then multiplied by an invertible n×n key matrix (modulo the alphabet size). For example, in a 3-letter
Hill cipher we treat “ACT” as a vector and multiply by a 3×3 matrix to encrypt it 4 . In general, each block of
n letters (as an n-vector) is multiplied by an invertible n×n key matrix (mod 26) 4 . This produces a
new vector of numbers, which we map back to letters to form the ciphertext. Crucially, the key matrix must
be invertible (its determinant nonzero and relatively prime to 26) so that one can compute its inverse. To
decrypt, each ciphertext vector is multiplied by the inverse of the key matrix, recovering the original
plaintext numbers 4 5 . In other words, encryption is

C=KP (mod 26),

and decryption is

P = K −1 C (mod 26),

where $\mathbf{P}$ is the plaintext vector, $\mathbf{C}$ is the ciphertext vector, and $K$ is the key matrix.
(Letters are often mapped to numbers A=0, B=1,…,Z=25, though other conventions like A=1,…,Z=26 can be
used.) In summary, the Hill cipher demonstrates how linear algebra (matrix multiplication and
inversion) can encode and decode messages 4 6 .

1
Real-World Applications

Matrix ciphers like the Hill cipher are mainly educational today, but the idea of using linear transformations
appears in modern cryptography. For instance, block ciphers (like AES) operate on data in fixed-size blocks
and use mathematical transformations that include matrix operations. The AES (Advanced Encryption
Standard) algorithm, for example, represents data as a 4×4 byte matrix and uses a fixed matrix
multiplication (the “MixColumns” step) to diffuse bits 7 . (A detailed analysis notes that “AES encryption
implements extensive algorithms… to manipulate matrices” 8 .)

In practice, encryption is everywhere in modern life: securing credit-card transactions, online banking,
email, messaging, and government communications. A survey of cryptography notes that practical
applications include electronic commerce, payment cards, digital currencies, passwords, and military
communication 9 . In each case, data is transformed (encrypted) so that only authorized parties can read
it. While modern ciphers often rely more on number theory (like RSA or elliptic curves) than simple matrices,
the Hill cipher was an important early example of “diffusion” – spreading the influence of each plaintext
letter across the block 10 . (In fact, with a large key matrix the Hill cipher “achieved Shannon’s diffusion” for
n letters at once 10 .)

It is important to note that the Hill cipher by itself is not secure by today’s standards. As one text
explains, it is “not considered a secure encryption method; it is relatively easy to break with modern
technology” 11 . Nevertheless, it introduced modern algebraic ideas into cryptography 11 . Today’s
encryption schemes are much more complex (combining substitution, permutation, and large prime math),
but the basic lesson remains: mathematics (often matrices or finite-field arithmetic) underlies the encoding
of secret messages 3 8 .

Activity Sheet
Below is a worked example of encoding and decoding a short message using a matrix. We will use the
mapping A=1, B=2, …, Z=26, and (space)=27 12 :

Letter A B C … Y Z (space)

Number 1 2 3 … 25 26 27

Step 1: Choose a message and key. Let the plaintext be “HELLO ” (with a space as the 6th character). We
choose a 2×2 invertible key matrix:

1 2
K=( ).
3 7

Its determinant is $1\cdot7 - 2\cdot3 = 1$, so $K$ is invertible (and in fact $K^{-1} = \begin{pmatrix}7 & -2 \
-3 & 1\end{pmatrix}$ over the integers).

Step 2: Convert letters to numbers and form vectors. Group the message in pairs of letters: “HE”, “LL”,
“O ( )”. Using the mapping, H=8, E=5, L=12, O=15, (space)=27. Thus the letter blocks become column vectors:

2
 8 12 15
p1 = ( ) , p2 = ( ), p3 = ( ).
5 12 27

Step 3: Encrypt by matrix multiplication. Multiply each plaintext vector by $K$ and reduce modulo 27 (so
results stay in the 1–27 range). For each block:

• For “HE” ($\mathbf{p_1}$):

1 2 8 1⋅8+2⋅5 8 + 10 18
Kp1 = ( )( ) = ( )=( ) = ( ).
3 7 5 3⋅8+7⋅5 24 + 35 59

Reducing each entry mod 27: $18 \equiv 18,\; 59 \equiv 5 \pmod{27}$. Thus the encrypted vector is $
\begin{pmatrix}18\5\end{pmatrix}$. Converting back to letters (using the same mapping) gives
$(18,5)\to\,$R, E.

• For “LL” ($\mathbf{p_2}$):

1 2 12 12 + 24 36
Kp2 = ( )( ) = ( )=( ).
3 7 12 36 + 84 120

Modulo 27: $36 \equiv 9,\; 120 \equiv 12 \pmod{27}$. So this yields $\begin{pmatrix}
9\12\end{pmatrix}$, i.e. I, L.

• For “O ( )” ($\mathbf{p_3}$):

1 2 15 15 + 54 69
Kp3 = ( )( ) = ( )=( ).
3 7 27 45 + 189 234

Mod 27: $69 \equiv 15,\; 234 \equiv 18$. So the result is $\begin{pmatrix}15\18\end{pmatrix}$, i.e. O,
R.

Putting it all together, the ciphertext vectors are $\begin{pmatrix}18\5\end{pmatrix}$, $\begin{pmatrix}

9\12\end{pmatrix}$, $\begin{pmatrix}15\18\end{pmatrix}$, which correspond to the letters “RE IL OR” (or
concatenated “REILOR”).

Step 4: Decode using the inverse matrix. We find $K^{-1} = \begin{pmatrix}7 & -2\ -3 & 1\end{pmatrix}$
(since $\det K=1$, the inverse has integer entries). To decrypt, multiply each ciphertext vector by $K^{-1}$
mod 27:

• Decode “RE” ($\mathbf{c_1} = \begin{pmatrix}18\5\end{pmatrix}$):

7 −2 18 7 ⋅ 18 − 2 ⋅ 5 126 − 10 116
K −1 c1 = ( )( ) = ( )=( )=( ).
−3 1 5 −3 ⋅ 18 + 1 ⋅ 5 −54 + 5 −49

3
 Mod 27: $116 \equiv 8,\; -49 \equiv 5 \pmod{27}$ (note $-49+2\cdot27 = 5$). This yields $
\begin{pmatrix}8\5\end{pmatrix}$ = H, E.

• Decode “IL” ($\mathbf{c_2} = \begin{pmatrix}9\12\end{pmatrix}$):

7 ⋅ 9 − 2 ⋅ 12 63 − 24 39
K −1 c2 = ( )=( )=( ).
−3 ⋅ 9 + 1 ⋅ 12 −27 + 12 −15

Mod 27: $39 \equiv 12,\; -15 \equiv 12$. So $\begin{pmatrix}12\12\end{pmatrix}$ = L, L.

• Decode “OR” ($\mathbf{c_3} = \begin{pmatrix}15\18\end{pmatrix}$):

7 ⋅ 15 − 2 ⋅ 18 105 − 36 69
K −1 c3 = ( )=( )=( ).
−3 ⋅ 15 + 1 ⋅ 18 −45 + 18 −27

Mod 27: $69 \equiv 15,\; -27 \equiv 0 \equiv 27$ (we treat 0 as 27). So we get $\begin{pmatrix}
15\27\end{pmatrix}$ = O, (space).

In each case we recovered the original number pair. Converting these back to letters yields “H E”, “L L”,
“O (space)”, i.e. “HELLO ”. Thus the decryption worked correctly.

Step-by-step Computations: In summary, the key computations were:

• Encryption of the first block: $K[8;5] = [18;59] \equiv [18;5]\ (\bmod 27)$.
• Encryption of the second block: $K[12;12] = [36;120] \equiv [9;12]\ (\bmod 27)$.
• Encryption of the third block: $K[15;27] = [69;234] \equiv [15;18]\ (\bmod 27)$.
• Inverse matrix: $\det K=1$, so

1 7 −2
K −1 = ( ).
1 −3 1

• Decryption of first block: $K^{-1}[18;5] = [116;-49] \equiv [8;5]\ (\bmod 27)$.

• Decryption of second block: $K^{-1}[9;12] = [39;-15] \equiv [12;12]\ (\bmod 27)$.
• Decryption of third block: $K^{-1}[15;18] = [69;-27] \equiv [15;27]\ (\bmod 27)$.

Every step used straightforward matrix multiplication and reduction, demonstrating how the encoded
message is obtained and then recovered by multiplying with the inverse.

Inferences and Conclusions

From this activity we learned that matrix multiplication can systematically encode blocks of text. Each
pair of letters was treated as a vector and multiplied by a key matrix, “mixing” the letters according to the
matrix entries. The encoded message looked unrelated to the plaintext (for example, “HELLO ” became “RE
IL OR”), illustrating how the matrix scrambles the letters. Crucially, because the key matrix was invertible,
we could reverse the process: multiplying the ciphertext by the inverse matrix exactly recovered the original

4
numbers and hence the original letters. This shows that invertibility is essential – if the matrix were not
invertible (determinant 0 or not coprime with 27), we could not solve back for the plaintext. In doing the
computations we reinforced skills in matrix multiplication, modular arithmetic, and finding inverses. Overall,
we observed that encoding with a matrix is just a change of basis in a finite vector space: it is fully reversible
if the key is known. The project underscored the connection between linear algebra and cryptography:
matrices provide a simple model of a cipher, and the same principles underlie more complex encryption
systems used in real life (though those use larger matrices or fields). Thus we saw how matrices can hide
and then reveal information, which is the core idea of encryption.

Bibliography
Sources used include educational texts and articles on matrix ciphers and cryptology:

• Sekhon, R. & Bloom, R., “Application of Matrices in Cryptography”, LibreTexts (2016) 13 6 – explains
Hill’s method and the general procedure of encoding with matrices.
• Wikipedia contributors, “Cryptography” (Wikipedia) 1 9 – definition of cryptography/cryptology
and its modern uses.
• Wikipedia contributors, “Hill cipher” (Wikipedia) 4 5 – describes the Hill cipher, including the
process of encryption and decryption using matrix multiplication.
• Awati, R., “What is cryptology?” (TechTarget, 2025) 2 – overview of cryptology as the mathematical
foundation of cryptography and cryptanalysis.
• Bogdanov, A. & Khovratovich, D., AES Encryption Analysis (from [Link]) 8 – discusses how
modern ciphers like AES use matrix operations.

Each source above was used to understand and explain how matrices encode/decode messages and how
these ideas apply in real-world cryptography.

1 3 9 Cryptography - Wikipedia
[Link]

2 What is Cryptology? | Definition from TechTarget

[Link]

4 5 10 Hill cipher - Wikipedia

[Link]

6 11 12 13 2.5: Application of Matrices in Cryptography - Mathematics LibreTexts

[Link]
02%3A_Matrices/2.05%3A_Application_of_Matrices_in_Cryptography

7 Rijndael MixColumns - Wikipedia

[Link]

8 [Link]
[Link]