• Aucun résultat trouvé

February11 ,2021 MODULARARITHMETIC ConceptsinAbstractMathematics

N/A
N/A
Protected

Academic year: 2022

Partager "February11 ,2021 MODULARARITHMETIC ConceptsinAbstractMathematics"

Copied!
10
0
0

Texte intégral

(1)

MAT246H1-S – LEC0201/9201

Concepts in Abstract Mathematics

MODULAR ARITHMETIC

February 11th, 2021

(2)

• This new layer of abstraction allowed to simplify previous proofs and to prove new theorems.

• Informally, we windℤon itself as below:

−13−12−11−10 −9 −8 −7 −6 −5 −4 −3 −2 −1 0 1 2 3 4 5 6 7 8 9 10 11 12 13

… , −12, −6, 0, 6, 12, …

… , −11, −5, 1, 7, 13, …

… , −10, −4, 2, 8, 14, …

… , −9, −3, 3, 9, 15, …

… , −8, −2, 4, 10, 16, … … , −7, −1, 5, 11, 17, …

ℤmodulo6

(3)

Congruences – 1

Definition: equivalence relation

We say that a binary relationℛon a set𝐸is anequivalence relationif

1 ∀𝑥 ∈ 𝐸, 𝑥ℛ𝑥 (reflexivity)

2 ∀𝑥, 𝑦 ∈ 𝐸, 𝑥ℛ𝑦 ⟹ 𝑦ℛ𝑥 (symmetry)

3 ∀𝑥, 𝑦, 𝑧 ∈ 𝐸, (𝑥ℛ𝑦and𝑦ℛ𝑧) ⟹ 𝑥ℛ𝑧 (transitivity)

(4)

We say that𝑎and𝑏arecongruent modulo𝑛, denoted by𝑎 ≡ 𝑏 (mod𝑛), if𝑛|𝑎 − 𝑏.

Proposition

Congruence modulo𝑛is an equivalence relation onℤ.

Proof.

Reflexivity. Let𝑎 ∈ ℤthen𝑛|0 = 𝑎 − 𝑎. Hence𝑎 ≡ 𝑎 (mod𝑛).

Symmetry. Let𝑎, 𝑏 ∈ ℤbe such that𝑎 ≡ 𝑏 (mod𝑛).

Then𝑛|𝑏 − 𝑎 = −(𝑎 − 𝑏)hence𝑏 ≡ 𝑎 (mod𝑛).

Transitivity.Let𝑎, 𝑏, 𝑐 ∈ ℤbe such that𝑎 ≡ 𝑏 (mod𝑛)and𝑏 ≡ 𝑐 (mod𝑛).

Then𝑛|𝑎 − 𝑏and𝑛|𝑏 − 𝑐. Hence𝑛|𝑎 − 𝑐 = (𝑎 − 𝑏) + (𝑏 − 𝑐).

Thus𝑎 ≡ 𝑐 (mod𝑛). ■

(5)

Congruences – 3

Proposition

Let𝑛 ∈ ℕ ⧵ {0}and𝑎, 𝑏 ∈ ℤ.

Then𝑎 ≡ 𝑏 (mod𝑛)if and only if𝑎and𝑏have same remainder for the Euclidean division by𝑛.

Proof.

⇒. Assume that𝑎 ≡ 𝑏 (mod𝑛), then𝑏 − 𝑎 = 𝑘𝑛for some𝑘 ∈ ℤ.

By Euclidean division,𝑎 = 𝑛𝑞 + 𝑟for𝑞, 𝑟 ∈ ℤsatisfying0 ≤ 𝑟 < 𝑛.

Hence𝑏 = 𝑎 + 𝑘𝑛 = 𝑛𝑞 + 𝑟 + 𝑘𝑛 = (𝑞 + 𝑘)𝑛 + 𝑟.

⇐. Assume that𝑎and𝑏have same remainder for the Euclidean division by𝑛.

Then𝑎 = 𝑛𝑞1+ 𝑟and𝑏 = 𝑛𝑞2+ 𝑟where𝑞1, 𝑞2, 𝑟 ∈ ℤwith0 ≤ 𝑟 < 𝑛.

Hence𝑎 − 𝑏 = 𝑛𝑞1+ 𝑟 − (𝑛𝑞2+ 𝑟) = 𝑛(𝑞1− 𝑞2).

Thus𝑛|𝑎 − 𝑏, i.e. 𝑎 ≡ 𝑏 (mod𝑛). ■

Corollary

Let𝑛 ∈ ℕ ⧵ {0}and𝑎 ∈ ℤ. Then𝑎is congruent modulo𝑛to exactly one element of{0, 1, … , 𝑛 − 1}.

(6)

• 𝑎𝑐 ≡ 𝑏𝑑 (mod𝑛)

Proof.Let𝑎, 𝑏, 𝑐, 𝑑 ∈ ℤand𝑛 ∈ ℕ ⧵ {0}. Assume that𝑎 ≡ 𝑏 (mod𝑛)and that𝑐 ≡ 𝑑 (mod𝑛).

Hence𝑎 − 𝑏 = 𝑛𝑘and𝑐 − 𝑑 = 𝑛𝑙for some𝑘, 𝑙 ∈ ℤ. Then

• (𝑎 + 𝑐) − (𝑏 + 𝑑) = (𝑎 − 𝑏) + (𝑐 − 𝑑) = 𝑛𝑘 + 𝑛𝑙 = 𝑛(𝑘 + 𝑙), hence𝑎 + 𝑐 ≡ 𝑏 + 𝑑 (mod𝑛).

• 𝑎𝑐 − 𝑏𝑑 = (𝑏 + 𝑛𝑘)(𝑑 + 𝑛𝑙) − 𝑏𝑑 = 𝑏𝑛𝑙 + 𝑑𝑛𝑘 + 𝑛2𝑘𝑙 = 𝑛(𝑏𝑙 + 𝑑𝑘 + 𝑛𝑘𝑙), hence𝑎𝑐 ≡ 𝑏𝑑 (mod𝑛). ■

Corollary

Let𝑎, 𝑏 ∈ ℤand𝑛 ∈ ℕ ⧵ {0}. Then∀𝑘 ∈ ℕ, 𝑎 ≡ 𝑏 (mod𝑛) ⟹ 𝑎𝑘≡ 𝑏𝑘(mod𝑛).

Proof.We prove the statement by induction on𝑘.

Base case at𝑘 = 0:𝑎0= 𝑏0= 1hence𝑎0≡ 𝑏0(mod𝑛).

Induction step:assume that𝑎 ≡ 𝑏 (mod𝑛) ⟹ 𝑎𝑘≡ 𝑏𝑘(mod𝑛)for some𝑘 ∈ ℕ.

If𝑎 ≡ 𝑏 (mod𝑛)then by the IH we also have𝑎𝑘≡ 𝑏𝑘(mod𝑛). Hence𝑎𝑘𝑎 ≡ 𝑏𝑘𝑏 (mod𝑛). ■

(7)

Modular multiplicative inverse

Proposition

Let𝑎 ∈ ℤand𝑛 ∈ ℕ ⧵ {0}. Then𝑎has a multiplicative inverse modulo𝑛if and only ifgcd(𝑎, 𝑛) = 1.

Otherwise stated,

∃𝑏 ∈ ℤ, 𝑎𝑏 ≡ 1 (mod𝑛) ⇔gcd(𝑎, 𝑛) = 1

Proof.∃𝑏 ∈ ℤ, 𝑎𝑏 ≡ 1 (mod𝑛) ⇔ ∃𝑏, 𝑐 ∈ ℤ, 𝑎𝑏 + 𝑛𝑐 = 1 ⇔gcd(𝑎, 𝑛) = 1 ■

Remark

When it exists, the multiplicative inverse is unique modulo𝑛.

Indeed, assume that𝑎𝑏 ≡ 1 (mod𝑛)and𝑎𝑏≡ 1 (mod𝑛)then𝑎𝑏 ≡ 𝑎𝑏(mod𝑛)so𝑛|𝑎(𝑏 − 𝑏).

Sincegcd(𝑎, 𝑛) = 1, by Gauss’ lemma we get𝑛|𝑏 − 𝑏. Therefore𝑏≡ 𝑏 (mod𝑛).

Note thatgcd(4, 25) = 1so4has a multiplicative inverse modulo25.

We may find one representative of the inverse from a Bézout’s identity: 4 × (−6) + 25 × 1 = 1.

So4 × (−6) ≡ 1 (mod25).

(8)

Proof.Note that10 ≡ 1 (mod3), hence 𝑎𝑟𝑎𝑟−1… 𝑎010=

𝑟

𝑘=0

𝑎𝑘10𝑘

𝑟

𝑘=0

𝑎𝑘1𝑘(mod3) ≡

𝑟

𝑘=0

𝑎𝑘(mod3)

Thus, 3|𝑎𝑟𝑎𝑟−1… 𝑎010⇔ 𝑎𝑟𝑎𝑟−1… 𝑎010≡ 0 (mod3) ⇔

𝑟

𝑘=0

𝑎𝑘≡ 0 (mod3) ⇔ 3|

𝑟

𝑘=0

𝑎𝑘

Examples

• 91524is divisible by3since9 + 1 + 5 + 2 + 4 = 21 = 7 × 3is.

• Let’s study whether8546921469is a multiple of3or not:

3|8546921469 ⇔ 3|8 + 5 + 4 + 6 + 9 + 2 + 1 + 4 + 6 + 9 = 54 ⇔ 3|5 + 4 = 9.

But9 = 3 × 3, hence3|8546921469.

(9)

Application: divisibility criterion for 9

Note that10 ≡ 1 (mod9), hence we have a similar result:

Proposition

9|𝑎𝑟𝑎𝑟−1… 𝑎010if and only if9|

𝑟

𝑘=0

𝑎𝑘

(10)

Proof.Note that10 = 4 × 25hence10 ≡ 0 (mod4)for𝑘 ≥ 2. Hence 4|𝑎𝑟𝑎𝑟−1… 𝑎010⇔ 𝑎𝑟𝑎𝑟−1… 𝑎010≡ 0 (mod4)

𝑟

𝑘=0

𝑎𝑘10𝑘≡ 0 (mod4)

⇔ 𝑎1× 10 + 𝑎0≡ 0 (mod4)

⇔ 𝑎1𝑎010≡ 0 (mod4)

⇔ 4|𝑎1𝑎010

Example

• 4 ∤ 856987454251100125since4 ∤ 25.

• 4|98854558715580since4|80 = 4 × 20.

Références

Documents relatifs

Proposition: the converse of Bézout’s identity holds for coprime numbers..

Prime numbers play a crucial role in number theory: they are the integers greater than 1 which can’t be factorized further while any other positive integer admits a unique expression

[r]

Note that both versions of Fermat’s little theorem

WILSON’S THEOREM &amp; THE CHINESE

[r]

How can someone send a secret message in a way that only the recipient could read the content even if the message happens to have been intercepted by a third party..

We usually generate prime numbers as follows (it is a little bit tricky):. 1 Generate a random odd