ℤₚ × ℤₚ Field

Problem:

∀p∈P,Zp×Zp={(a,b):a,b∈Zp} \forall p \in \mathbb{P}, \mathbb{Z}_{p} \times \mathbb{Z}_{p}= \{(a,b):a,b \in Z_{p} \}

Does (Zp×Zp,+,.)(\mathbb{Z}_{p} \times \mathbb{Z}p,+,.) define a field? If so, how should its addition and multiplication operations be defined?

Let’s look at Zp\mathbb{Z}_{p}.
(Zp,+)=(Z/pZ,+)(\mathbb{Z}_{p},+)=(\mathbb{Z}/p \mathbb{Z},+) defines a cyclic group under modular addition, and (Zp∗,.)(\mathbb{Z}_{p}^{*},.) is a group under modular multiplication, where Zp∗={a∈Zp:gcd(a,p)=1}\mathbb{Z}_{p}^{*}= \{a \in \mathbb{Z}_{p}:gcd(a,p)=1 \} and ∣Zp∗∣=φ(p)| \mathbb{Z}_{p}^{*}|= \varphi(p) (in this case, the integers from 1 to p-1).
We also know that (Zp,+,.)(\mathbb{Z}_{p},+,.) is a field, unlike (Z,+,.)(\mathbb{Z},+,.), which is only an integral domain; this is easy to prove.

Now let’s look at polynomial rings.
Theorem 1.1: (R[x],+,.)(R[x],+,.), the polynomial ring over the integral domain RR, defines an integral domain.

xn=(...,0,0,1‾,0,0,...) x^{n}=(...,0,0, \underline{1},0,0,...)

R[x]={Σn=0∞anxn:an∈R} R[x]= \{\Sigma_{n=0}^{\infty}a_{n}x^{n}:a_{n} \in R \}

(a0+a1x+...+anxn)+(b0+b1x1+...+bmxm)=(a0+b0)+(a1+b1)x+... (a_{0}+a_{1}x+...+a_{n}x^{n})+(b_{0}+b_{1}x^{1}+...+b_{m}x^{m})=(a_{0}+b_{0})+(a_{1}+b_{1})x+...

(a0+a1x+...+anxn).(b0+b1x1+...+bmxm)=c0+c1x+...+cn+mxn+m (a_{0}+a_{1}x+...+a_{n}x^{n}).(b_{0}+b_{1}x^{1}+...+b_{m}x^{m})=c_{0}+c_{1}x+...+c_{n+m}x^{n+m}

where ck=Σi=0kaibk−ic_{k}= \Sigma_{i=0}^{k}a_{i}b_{k-i}.
Proof.

  • unity: ∀f(x)∈R[x],(1x0).f(x)=f(x).(1x0)=f(x)\forall f(x) \in R[x],(1x^{0}).f(x)=f(x).(1x^{0})=f(x)
  • commutative ring: ck=Σi=0kaibk−i=Σi=0kbiak−ic_{k}= \Sigma_{i=0}^{k}a_{i}b_{k-i}= \Sigma_{i=0}^{k}b_{i}a_{k-i}
  • no nonzero zero divisors:
    This means f(x).g(x)=0⟶f(x)=0∨g(x)=0f(x).g(x)=0 \longrightarrow f(x)=0 \vee g(x)=0, since we have ck=Σi=0kaibk−ic_{k}= \Sigma_{i=0}^{k}a_{i}b_{k-i} and RR is an integral domain, ckc_{k} would not be zero if f(x)f(x) and g(x)g(x) are both non-null polynomials. □\square

Definition (Maximal Ideal): An ideal II is a maximal ideal of (R,+,.)(R,+,.) iff there is no ideal I1I_{1} in RR such that I⊂I1⊂RI \subset I_{1} \subset R.
Theorem 1.2: Maximal ideals of F[x]F[x] are of the form ⟨f(x)⟩={p(x)∈F[x]:∃g(x)∈F[x],p(x)=f(x)g(x)}\langle f(x) \rangle= \{p(x) \in F[x]: \exists g(x) \in F[x],p(x)=f(x)g(x) \} where f(x)f(x) is an irreducible polynomial in F[x]F[x].
Proof.
Proof by contradiction. Assume f(x)f(x) is reducible and ⟨f(x)⟩\langle f(x) \rangle is a maximal ideal of F[x].
So ∃g(x),t(x)∈F[x],p(x)=g(x).t(x)\exists g(x),t(x) \in F[x],p(x)=g(x).t(x). This means ⟨f(x)⟩⊂⟨g(x)⟩∧⟨f(x)⟩⊂⟨t(x)⟩\langle f(x) \rangle \subset \langle g(x) \rangle \wedge \langle f(x) \rangle \subset \langle t(x) \rangle.
But we assumed that ⟨f(x)⟩\langle f(x) \rangle is a maximal ideal of F[x]. ⊥\bot

Theorem 1.3: If II is a maximal ideal and (R,+,.)(R,+,.) is a commutative ring with unity, then the factor ring R/IR/I is a field.

Theorem 1.4: Let II be an ideal of (R,+,.)(R,+,.), then (R/I,+′,.′)(R/I,+^{'},.^{'}) is a ring.
     (a+I)+′(b+I)=(a+b)+I(a+I)+^{'}(b+I)=(a+b)+I
     (a+I).′(b+I)=(a.b)+I(a+I).^{'}(b+I)=(a.b)+I

Example: By Theorem 1.1, Z3[x]\mathbb{Z}_{3}[x] is an integral domain. Let f(x)=x2+1f(x)=x^{2}+1; it is irreducible in Z[x]\mathbb{Z}[x].
Z3[x]/⟨f(x)⟩={⟨f(x)⟩,(1)+⟨f(x)⟩,(2)+⟨f(x)⟩,(x)+⟨f(x)⟩,(x+1)+⟨f(x)⟩,(x+2)+⟨f(x)⟩,(2x)+⟨f(x)⟩,(2x+1)+⟨f(x)⟩,(2x+2)+⟨f(x)⟩}\mathbb{Z}_{3}[x]/ \langle f(x) \rangle= \{\langle f(x) \rangle,(1)+ \langle f(x) \rangle,(2)+ \langle f(x) \rangle,(x)+ \langle f(x) \rangle,(x+1)+ \langle f(x) \rangle,(x+2)+ \langle f(x) \rangle,(2x)+ \langle f(x) \rangle,(2x+1)+ \langle f(x) \rangle,(2x+2)+ \langle f(x) \rangle \}
∣Z3[x]/⟨f(x)⟩∣=9| \mathbb{Z}_{3}[x]/ \langle f(x) \rangle|=9

Definition (Homomorphism): A map f:(R1,+1,.1)→(R2,+2,.2)f:(R_{1},+_{1},._{1}) \to(R_{2},+_{2},._{2}) is a ring homomorphism if ∀x,y∈R1,f(x+1y)=f(x)+2f(y)\forall x,y \in R_{1},f(x+_{1}y)=f(x)+_{2}f(y) and f(x.1y)=f(x).2f(y)f(x._{1}y)=f(x)._{2}f(y).
Definition (Isomorphism): A homomorphism f:(R1,+1,.1)→(R2,+2,.2)f:(R_{1},+_{1},._{1}) \to(R_{2},+_{2},._{2}) is a ring isomorphism if ff is a one-to-one correspondence.
Theorem 1.5: Let f:(R,+1,.1)→(F,+2,.2)f:(R,+_{1},._{1}) \to(F,+_{2},._{2}) be an isomorphism. If F is a field, then R is also a field.
Theorem 1.6: Let f:(R1,+1,.1)→(R2,+2,.2)f:(R_{1},+_{1},._{1}) \to(R_{2},+_{2},._{2}) be a homomorphism; then ker(f)={x∈R1:f(x)=02}ker(f)= \{x \in R_{1}:f(x)=0_{2} \} is an ideal of R1R_{1}.
Theorem 1.7 (First Isomorphism Theorem): Let f:(R1,+1,.1)→(R2,+2,.2)f:(R_{1},+_{1},._{1}) \to(R_{2},+_{2},._{2}) be a homomorphism; then h:R1/ker(f)≅Im(f)h:R_{1}/ker(f) \cong Im(f) with the rule h(a+ker(f))=f(a)h(a+ker(f))=f(a).
Definition (Degree): deg(f(x))deg(f(x)) is the largest n∈Wn \in \mathbb{W} such that an≠0a_{n} \ne0.
Theorem 1.8 (Euclid’s Division Lemma): Let f(x)≠0,g(x)∈F[x]f(x) \ne0,g(x) \in F[x], then ∃q(x),r(x)∈F[x],g(x)=f(x).q(x)+r(x)\exists q(x),r(x) \in F[x],g(x)=f(x).q(x)+r(x) such that 0≤deg(r(x))<deg(f(x))0 \le deg(r(x))<deg(f(x))

Definition (Root): ω\omega is called a root of the polynomial f(x) iff f(ω)=0f(\omega)=0 i.e. ∃g(x)∈F[x],f(x)=(x−ω)g(x)\exists g(x) \in F[x],f(x)=(x-\omega)g(x)
Example: Let f(x)=(−8+11x+−6x2+x3)∈Z[x]f(x)=(-8+11x+-6x^{2}+x^{3}) \in \mathbb{Z}[x], then ω1=1\omega_{1}=1, ω2=2\omega_{2}=2 and ω3=3\omega_{3}=3.

So, let’s play the game.
Main Theorem: Let (R,+,.)(R,+,.) be an integral domain and p(x)p(x) be an irreducible polynomial in (R[x],+,.)(R[x],+,.) such that p(ω)=0p(\omega)=0 then R[x]/⟨p(x)⟩≅F[ω]R[x]/ \langle p(x) \rangle \cong F[\omega] and F[ω]F[\omega] is a field.
Proof.
     Let f:(R[x],+,.)→(R[ω],+,.)f:(R[x],+,.) \to(R[\omega],+,.) be a ring homomorphism.
     f(a0+a1x+a2x2+...+anxn)=a0+a1ω+a2ω2+...+anωnf(a_{0}+a_{1}x+a_{2}x^{2}+...+a_{n}x^{n})=a_{0}+a_{1} \omega+a_{2} \omega^{2}+...+a_{n} \omega^{n}
     By Theorem 1.6, ker(f)={q(x)∈R[x]:f(q(x))=0}ker(f)= \{q(x) \in R[x]:f(q(x))=0 \}.
     By Theorem 1.8, ∀g(x)∈R[x],∃t(x),r(x)∈R[x],g(x)=p(x).t(x)+r(x)\forall g(x) \in R[x], \exists t(x),r(x) \in R[x],g(x)=p(x).t(x)+r(x)
                    f(g(x))=f(p(x)).f(t(x))+f(r(x))f(g(x))=f(p(x)).f(t(x))+f(r(x)) by the definition of a homomorphism
                    g(ω)=p(ω).t(ω)+r(ω)g(\omega)=p(\omega).t(\omega)+r(\omega)
                    By assumption, p(ω)=0p(\omega)=0 then g(ω)=0.t(ω)+r(ω)g(\omega)=0.t(\omega)+r(\omega)
                    g(ω)=r(ω)g(\omega)=r(\omega)
                    This means f(g(x))=0⟷r(x)=0f(g(x))=0 \longleftrightarrow r(x)=0
     So ker(f)={q(x)∈R[x]:∃t(x)∈R[x],q(x)=p(x)t(x)}=⟨p(x)⟩ker(f)= \{q(x) \in R[x]: \exists t(x) \in R[x],q(x)=p(x)t(x) \}= \langle p(x) \rangle
     By Theorem 1.7, R[x]/⟨p(x)⟩≅Im(f)=f(R[x])=R[ω]R[x]/ \langle p(x) \rangle \cong Im(f)=f(R[x])=R[\omega]
     By Theorem 1.1, R[x]R[x] is an integral domain and by Theorem 1.3 R[x]/⟨p(x)⟩R[x]/ \langle p(x) \rangle is a field.
     By Theorem 1.5, R[ω]R[\omega] is also a field. □\square

It’s becoming interesting…

Example: (Z5,+,.)(\mathbb{Z}_{5},+,.) is a field. Let p(x)=x2−3p(x)=x^{2}-3 be an irreducible polynomial in Z5[x]\mathbb{Z}_{5}[x] with ω2=3\omega^{2}=3. By applying the Main Theorem, Z5[x]/⟨x2−3⟩≅Z5[ω]\mathbb{Z}_{5}[x]/ \langle x^{2}-3 \rangle \cong \mathbb{Z}_{5}[\omega].
Z5[x]/⟨x2−3⟩={(a+bx)+⟨x2−3⟩:a,b∈Z5}\mathbb{Z}_{5}[x]/ \langle x^{2}-3 \rangle= \{(a+bx)+ \langle x^{2}-3 \rangle:a,b \in \mathbb{Z}_{5} \}
Z5[ω]={a+bω:a,b∈Z5}\mathbb{Z}_{5}[\omega]= \{a+b \omega:a,b \in \mathbb{Z}_{5} \}
∣Z5[x]/⟨x2−3⟩∣=∣Z5[ω]∣=25| \mathbb{Z}_{5}[x]/ \langle x^{2}-3 \rangle|=| \mathbb{Z}_{5}[\omega]|=25
This is because the number of permutations of a+bωa+b \omega is 25, since in Theorem 1.8 we have that 0≤deg(r(x))<deg(p(x))0 \le deg(r(x))<deg(p(x)).

So let’s generalize for Zp×Zp\mathbb{Z}_{p} \times \mathbb{Z}_{p}:
Zp\mathbb{Z}_{p} is a field; let p(x)p(x) be an irreducible polynomial in Zp[x]\mathbb{Z}_{p}[x] such that deg(p(x))=2deg(p(x))=2 and p(ω)=0p(\omega)=0.
Then, by the Main Theorem, Zp[x]/⟨p(x)⟩≅Zp[ω]\mathbb{Z}_{p}[x]/ \langle p(x) \rangle \cong \mathbb{Z}_{p}[\omega].
By Theorem 1.8, ∣Zp[ω]∣=p2=∣Zp×Zp∣| \mathbb{Z}_{p}[\omega]|=p^{2}=| \mathbb{Z}_{p} \times \mathbb{Z}_{p}|.
Let (a,b)=a+bω(a,b)=a+b \omega; then we can define addition and multiplication on Zp×Zp\mathbb{Z}_{p} \times \mathbb{Z}_{p}:
(a,b)+(c,d)=(a+bω)+(c+dω)=(a+c)+(b+d)ω=(a+c,b+d)(a,b)+(c,d)=(a+b \omega)+(c+d \omega)=(a+c)+(b+d) \omega=(a+c,b+d)
(a,b).(c,d)=(a+bω).(c+dω)=a.c+(a.d)ω+(b.c)ω+b.d.ω2=(a.c+b.d.ω2,a.d+b.c)(a,b).(c,d)=(a+b \omega).(c+d \omega)=a.c+(a.d) \omega+(b.c) \omega+b.d. \omega^{2}=(a.c+b.d. \omega^{2},a.d+b.c)

Question: How do we know that an irreducible polynomial in Zp[x]\mathbb{Z}_{p}[x] such that deg(p(x))=2deg(p(x))=2 exists?
Theorem 1.9: Let (R, +, .) be a ring; then ∀a,b∈R,(−a).b=a.(−b)=−(a.b)\forall a,b \in R,(-a).b=a.(-b)=-(a.b)
Theorem 1.10 (Lagrange’s Theorem): Let GG be a group and H≤GH \le G, then ∣G∣=[G:H].∣H∣|G|=[G:H].|H|.
Theorem: Let (Zp,+,.)(\mathbb{Z}_{p},+,.) be a field. Then there are at least ⌊p2⌋\lfloor \frac{p}{2} \rfloor irreducible polynomials of the form p(x)p(x) in Zp[x]\mathbb{Z}_{p}[x] such that deg(p(x))=2deg(p(x))=2.
Proof.

  • p=2p=2:
    There exists only x2+x+1x^{2}+x+1.
  • p>2p>2:
    Our polynomials must have no root in Zp\mathbb{Z}_{p}, because any polynomial with a root is divisible by (x−ω)(x-\omega), which makes it reducible.
    Let Zp∗={a∈Zp:gcd(a,p)=1}\mathbb{Z}_{p}^{*}= \{a \in \mathbb{Z}_{p}:gcd(a,p)=1 \}, so ∣Zp∗∣=φ(p)=p−1| \mathbb{Z}_{p}^{*}|= \varphi(p)=p-1
    Since the polynomials must have no root in Zp\mathbb{Z}_{p}, we remove the integers a∈Zp∗a \in \mathbb{Z}_{p}^{*} such that a∗a=ω2a*a= \omega^{2}.
    By Theorem 1.9, a∗a=(−a)(−a)=ω2a*a=(-a)(-a)= \omega^{2}. This means there are two different integers in Zp∗\mathbb{Z}_{p}^{*} that produce ω2\omega^{2} because there is no element of order 2 by Theorem 1.10.
    So there should be ⌊p2⌋ω2\lfloor \frac{p}{2} \rfloor \omega^{2}’s to choose from, and the polynomials are of the form (x2−ω2)(x^{2}-\omega^{2}). □\square

← all posts

▲