University of Benghazi, Libya
* Corresponding author
University of Benghazi, Libya

Article Main Content

In this paper, semidirect products are used to find the matrix representations of group algebras of non-abelian groups of order p3, for a prime p ≥ 3.

Preliminaries

Let G be a group, assume that H is a normal subgroup of G, K is a subgroup of G, HK={1}, and G=HK. Suppose that K acts on H by automorphisms of H, then there exists a homomorphism φ:KAut(H). Assume the action is by conjugation, then for kK and hH we have k.h=(k)(h)=khk1. G is an internal semidirect product of H and K by φ, it is denoted by G=HφK [1].

Non-abelian groups of orders p3, for a prime p3 are of two types [1]:

G 1 = C p 2 α φ C p β and G 2 = ( C p α × C p β ) φ C p γ .

Thus,

G 1 = α , β : α p 2 = β p = 1 , β α = α 1 + p β

and

G 2 = α , β , γ : α p = β p = γ p = 1 , α β = β α , γ β = β γ , γ α = α β γ

Let F be a field. A ring A with unity is an algebra over F (breifly F-algebra) if A is a vector space over F and the following compatibility condition holds (sa).b=s(a.b)=a.(sb) for any a,bA and any sF. A is also called associative algebra (over F). The dimension of the algebra A is the dimension of A as a vector space over F.

Theorem 1 [2]

Let A be a n-dimensional algebra over a field F. Then there is a one-to-one algebra homomorphism from A into Mn(F), the algebra of n-matrices over F.

Let G={g1=1,g2,,gn} be a finite group of order n and F a field. Define FG={a1g1+a2g2++angn:aiF}. FG is n-dimensional vector space over F with basis G. Multiplication of G can be extended linearly to FG. Thus, FG becomes an algebra over F of dimension n. FG is called group algebra. The following identifications should be realized:

  1. 0FgG=0FG=0 for any gG.
  2. 1FgG=gFG=g for any gG. In particular 1F1G=1FG=1.
  3. aF1G=aFG for any aF.

A circulant matrix M on parameters a0,a1,,an1 is defined as follows:

M ( a 0 , a 1 , , a n 1 ) = [ a 0 a n 1 a 1 a 0 a 1 a 2 a n 1 a n 2 a 0 ]

This matrix may be denoted in terms of its columns by [col(a0)|col(an1)||col(a1)].

M is said to be circulant block matrix if it is of the form M(M1,M2,,Mn). i.e., it is circulant blockwise on the blocks M1,M2,,Mn.

Thus,

M = [ M 1 M n M 2 M 1 M 2 M 3 M n M n 1 M 1 ] .

Main Results

Theorem 2 [3]

Let F be a field and G=α:αn=1 a cyclic group of order n. Then any element a01+a1α++an1αn1 of FG can be represented with respect to the ordered basis {1,α,,αn1} by the circulant matrix M(a0,a1,,an1).

Proof

Let w=a01+a1α++an1αn1 be in FG. wα=a0α+a1α2++an11=an11+a0α++an2αn1wαn1=a0αn1+a11++an1αn2=a11+a2α++a0αn1. Then the matrix representation of w with respect to the basis {1,α,,αn1} is [a0an1a1a0a1a2an1an2a0] which is M(a0,a1,,an1).

Note that if the order of the basis elements is changed, we obtain a different matrix of representation. The new matrix is obtained by suitable interchanging of the columns of the matrix M(a0,a1,,an1). In [4] the representation is done for the non-split metacyclic group.

For more complicated finite groups we use the circulant block matrices to do the required representations.

Now, let G be an internal semidirect product of H and a cyclic group K=γ by φ.

Then the matrix representation [w] of the general element w in FG is given as follows:

G=HφK, φ:KAut(H) is a homomorphism, φ(γ)(h)=γhγ1. Suppose that H={h1,h2,,hn}, K=Cmγ={1,γ,,γm1} then the general element w in FG is w=a1h11+a2h21++anhn1+an+1h1γ+an+2h2γ++a2nhnγ+a2n+1h1γ2++a3nhnγ2++amnhnγm1. Now we can write w as: where

w = w 1 + w 2 + + w m ,
w 1 = a 1 h 1 1 + a 2 h 2 1 + + a n h n 1
w 2 = a n + 1 h 1 γ + a n + 2 h 2 γ + + a 2 n h n γ
w m = a ( m 1 ) ( n + 1 ) h 1 γ m 1 + + a m n h n γ m 1

The matrix representation [w] of w is [w]=M([w1],[w2]γ,,[wm]γm1), where γi:HH is the automorphism γi= φ(γ)(h)=γihγi and [wi]=[col(h1)|col(h2)||col(hn)], [wi]γi=[col(γi(h1))|col(γi(h2))||col(γi(hn))|]. Thus, we get the following theorem:

Theorem 3

With the above notations, the matrix representation [w] of the general element w in FG.

[ w ] = [ [ w 1 ] [ w m ] γ m 1 [ w 2 ] γ [ w 2 ] γ [ w 1 ] [ w m ] γ 2 [ w m ] γ m 1 [ w m ] γ m 2 [ w 1 ] ] .

Applications

Finally, we use theorem 3 to compute the matrix representations of FG1 and FG2, when the prime p=3.

1) G1=α,β:α32=β3=1,βα=α1+3β

  ={1,α,α2,,α8,β,αβ,α2β,,α8β,β2,αβ2,α2β2,,α8β2}.

The general element of FG1 is w=a01+a1α++a8α8+a9β+a10αβ++a17α8β+a18β2+a19αβ2++a26α8β2. Let w1=a01+a1α++a8α8, w2=a9β+a10αβ++a17α8β, w3=a18β2+a19αβ2++a26α8β2. Then w=w1+w2+w3.

By theorem 3, matrix representation of w is [w]=[[w1][w3]β2[w2]β[w2]β[w1][w3]β2[w3]β2[w2]β[w1]]

[ w 1 ] = [ a 0 a 8 a 7 a 6 a 5 a 4 a 3 a 2 a 1 a 1 a 0 a 8 a 7 a 6 a 5 a 4 a 3 a 2 a 2 a 1 a 0 a 8 a 7 a 6 a 5 a 4 a 3 a 3 a 2 a 1 a 0 a 8 a 7 a 6 a 5 a 4 a 4 a 3 a 2 a 1 a 0 a 8 a 7 a 6 a 5 a 5 a 4 a 3 a 2 a 1 a 0 a 8 a 7 a 6 a 6 a 5 a 4 a 3 a 2 a 1 a 0 a 8 a 7 a 7 a 6 a 5 a 4 a 3 a 2 a 1 a 0 a 8 a 8 a 7 a 6 a 5 a 4 a 3 a 2 a 1 a 0 ]

G1=C9αφC3β, φ:C3βAut(C9α) is a homomorphism such that φ(β)(α)=βα3β

φ(β)(1)=β1β1=1, φ(β)(α)=βαβ1=α4, φ(β)(α2)=βα2β1=α8, φ(β)(α3)=βα3β=α3, φ(β)(α4)=βα4β1=α7, φ(β)(α5)=βα5β1=α2, φ(β)(α6)=βα6β1=α6, φ(β)(α7)=βα7β1=α, φ(β)(α8)=βα8β1=α5.

[ w 2 ] = [ c o l ( 1 ) | c o l ( α ) | c o l ( α 2 ) | c o l ( α 3 ) | c o l ( α 4 ) | c o l ( α 5 ) | c o l ( α 6 ) | c o l ( α 7 ) | c o l ( α 8 ) ]
[ w 2 ] β = [ c o l ( 1 ) | c o l ( α 4 ) | c o l ( α 8 ) | c o l ( α 3 ) | c o l ( α 7 ) | c o l ( α 2 ) | c o l ( α 6 ) | c o l ( α ) | c o l ( α 5 ) ]
[ w 2 ] β = [ a 9 a 14 a 10 | a 15 a 11 a 16 | a 12 a 17 a 13 a 10 a 15 a 11 | a 16 a 12 a 17 | a 13 a 9 a 14 a 11 a 16 a 12 | a 17 a 13 a 9 | a 14 a 10 a 15 a 12 a 17 a 13 | a 9 a 14 a 10 | a 15 a 11 a 16 a 13 a 9 a 14 | a 10 a 15 a 11 | a 16 a 12 a 17 a 14 a 10 a 15 | a 11 a 16 a 12 | a 17 a 13 a 9 a 15 a 11 a 16 | a 12 a 17 a 13 | a 9 a 14 a 10 a 16 a 12 a 17 | a 13 a 9 a 14 | a 10 a 15 a 11 a 17 a 13 a 9 | a 14 a 10 a 15 | a 11 a 16 a 12 ]
φ ( β 2 ) ( α ) = β 2 α β 2

φ(β2)(1)=β21β2=1, φ(β2)(α)=β2αβ2=α7, φ(β2)(α2)=β2α2β2=α5, φ(β2)(α3)=β2α3β2=α3, φ(β2)(α4)=β2α4β2=α,φ(β2)(α5)=β2α5β2=α8, φ(β6)(α6)=β2α6β2=α6, φ(β2)(α7)=β2α7β2=α4, φ(β2)(α8)=β2α8β2=α2.

[ w 3 ] = [ c o l ( 1 ) | c o l ( α ) | c o l ( α 2 ) | c o l ( α 3 ) | c o l ( α 4 ) | c o l ( α 5 ) | c o l ( α 6 ) | c o l ( α 7 ) | c o l ( α 8 ) ]
[ w 3 ] β 2 = [ c o l ( 1 ) | c o l ( α 7 ) | c o l ( α 5 ) | c o l ( α 3 ) | c o l ( α ) | c o l ( α 8 ) | c o l ( α 6 ) | c o l ( α 4 ) | c o l ( α 2 ) ]
[ w 3 ] β 2 = [ a 18 a 20 a 22 | a 24 a 26 a 19 | a 21 a 23 a 25 a 19 a 21 a 23 | a 25 a 18 a 20 | a 22 a 24 a 26 a 20 a 22 a 24 | a 26 a 19 a 21 | a 23 a 25 a 18 a 21 a 23 a 25 | a 18 a 20 a 22 | a 24 a 26 a 19 a 22 a 24 a 26 | a 19 a 21 a 23 | a 25 a 18 a 20 a 23 a 25 a 18 | a 20 a 22 a 24 | a 26 a 19 a 21 a 24 a 26 a 19 | a 21 a 23 a 25 | a 18 a 20 a 22 a 25 a 18 a 20 | a 22 a 24 a 26 | a 19 a 21 a 23 a 26 a 19 a 21 | a 23 a 25 a 18 | a 20 a 22 a 24 ]

2) G2=α,β,γ:α3=β3=γ3=1,αβ=βα,γβ=βγ,γα=αβγ G2={1,α,α2,β,αβ,α2β,β2,αβ2,α2β2,γ,αγ,α2γ,βγ,αβγ,α2βγ,β2γ,αβ2γ, α2β2γ,γ2,αγ2,α2γ2,βγ2,αβγ2,α2βγ2,β2γ2,αβ2γ2,α2β2γ2}

The general element of FG2 is w=a01+a1α+a2α2+a3β+a4αβ+a5α2β+a6β2+a7αβ2+a8α2β2+a9γ+a10αγ+a11α2γ+a12βγ+a13αβγ+a14α2βγ+a15β2γ+a16αβ2γ+a17α2β2γ+a18γ2+a19αγ2+a20α2γ2+a21βγ2+a22αβγ2+a23α2βγ2+a24β2γ2+a25αβ2γ2+a26α2β2γ2.

w 1 = a 0 1 + a 1 α + a 2 α 2 + a 3 β + a 4 α β + a 5 α 2 β + a 6 β 2 + a 7 α β 2 + a 8 α 2 β 2
w 2 = a 9 γ + a 10 α γ + a 11 α 2 γ + a 12 β γ + a 13 α β γ + a 14 α 2 β γ + a 15 β 2 γ + a 16 α β 2 γ
w 3 = a 18 γ 2 + a 19 α γ 2 + a 20 α 2 γ 2 + a 21 β γ 2 + a 22 α β γ 2 + a 23 α 2 β γ 2 + a 24 β 2 γ 2 + a 25 α β 2 γ 2 + a 26 α 2 β 2 γ 2 .

Then w=w1+w2+w3.

The matrix representation of w is [w]=[[w1][w3]γ2[w2]γ[w2]γ[w1][w3]γ2[w3]γ2[w2]γ[w1]]

[ w 1 ] = [ a 0 a 2 a 1 | a 6 a 8 a 7 | a 3 a 5 a 4 a 1 a 0 a 2 | a 7 a 6 a 8 | a 4 a 3 a 5 a 2 a 1 a 0 | a 8 a 7 a 6 | a 5 a 4 a 3 a 3 a 5 a 4 | a 0 a 2 a 1 | a 6 a 8 a 7 a 4 a 3 a 5 | a 1 a 0 a 2 | a 7 a 6 a 8 a 5 a 4 a 3 | a 2 a 1 a 0 | a 8 a 7 a 6 a 6 a 8 a 7 | a 3 a 5 a 4 | a 0 a 2 a 1 a 7 a 6 a 8 | a 4 a 3 a 5 | a 1 a 0 a 2 a 8 a 7 a 6 | a 5 a 4 a 3 | a 2 a 1 a 0 ]

G2=(C3α×C3β)φC3γ, φ:C3γAut(C3α×C3β) is a homomorphism such that φ(γ)(α)=γαγ1.

φ(γ)(1)=γ1γ1=1, φ(γ)(α)=γαγ1=αβ, φ(γ)(α2)=γα2γ1=α2β2, φ(γ)(β)=γβγ1=β, φ(γ)(αβ)=γαβγ1=αβ2, φ(γ)(α2β)=γα2βγ1=α2, φ(γ)(β2)=γβ2γ1=β2, φ(γ)(αβ2)=γαβ2γ1=α, φ(γ)(α2β2)=γα2β2γ1=α2β.

[ w 2 ] = [ c o l ( 1 ) | c o l ( α ) | c o l ( α 2 ) | c o l ( β ) | c o l ( α β ) | c o l ( α 2 β ) | c o l ( β 2 ) | c o l ( α β 2 ) | c o l ( α 2 β 2 ) ]
[ w 2 ] γ = [ c o l ( 1 ) | c o l ( α β ) | c o l ( α 2 β 2 ) | c o l ( β ) | c o l ( α β 2 ) | c o l ( α 2 ) | c o l ( β 2 ) | c o l ( α ) | c o l ( α 2 β ) ]
[ w 2 ] γ = [ a 9 a 17 a 13 | a 15 a 14 a 10 | a 12 a 11 a 16 a 10 a 15 a 14 | a 16 a 12 a 11 | a 13 a 9 a 17 a 11 a 16 a 12 | a 17 a 13 a 9 | a 14 a 10 a 15 a 12 a 11 a 16 | a 9 a 17 a 13 | a 15 a 14 a 10 a 13 a 9 a 17 | a 10 a 15 a 14 | a 16 a 12 a 11 a 14 a 10 a 15 | a 11 a 16 a 12 | a 17 a 13 a 9 a 15 a 14 a 10 | a 12 a 11 a 16 | a 9 a 17 a 13 a 16 a 12 a 11 | a 13 a 9 a 17 | a 10 a 15 a 14 a 17 a 13 a 9 | a 14 a 10 a 15 | a 11 a 16 a 12 ]
φ ( γ 2 ) ( α ) = γ 2 α γ 2

φ(γ2)(1)=γ21γ2=1, φ(γ2)(α)=γ2αγ2=αβ2, φ(γ2)(α2)=γ2α2γ2=α2β, φ(γ2)(β)=γ2βγ2=β, φ(γ2)(αβ)=γ2αβγ2=α, φ(γ2)(α2β)=γ2α2βγ2=α2β2, φ(γ2)(β2)=γ2β2γ2=β2, φ(γ2)(αβ2)=γ2αβ2γ2=αβ, φ(γ2)(α2β2)=γ2α2β2γ2=α2.

[ w 3 ] = [ c o l ( 1 ) | c o l ( α ) | c o l ( α 2 ) | c o l ( β ) | c o l ( α β ) | c o l ( α 2 β ) | c o l ( β 2 ) | c o l ( α β 2 ) | c o l ( α 2 β 2 ) ]
[ w 3 ] γ 2 = [ c o l ( 1 ) | c o l ( α β 2 ) | c o l ( α 2 β ) | c o l ( β ) | c o l ( α ) | c o l ( α 2 β 2 ) | c o l ( β 2 ) | c o l ( α β ) | c o l ( α 2 ) ]
[ w 3 ] γ 2 = [ a 18 a 23 a 25 | a 24 a 20 a 22 | a 21 a 26 a 19 a 19 a 21 a 26 | a 25 a 18 a 23 | a 22 a 24 a 20 a 20 a 22 a 24 | a 26 a 19 a 21 | a 23 a 25 a 18 a 21 a 26 a 19 | a 18 a 23 a 25 | a 24 a 20 a 22 a 22 a 24 a 20 | a 19 a 21 a 26 | a 25 a 18 a 23 a 23 a 25 a 18 | a 20 a 22 a 24 | a 26 a 19 a 21 a 24 a 20 a 22 | a 21 a 26 a 19 | a 18 a 23 a 25 a 25 a 18 a 23 | a 22 a 24 a 20 | a 19 a 21 a 26 a 26 a 19 a 21 | a 23 a 25 a 18 | a 20 a 22 a 24 ]

For greater prime p, the same method may be applied.

Conflict of Interest

Conflict of Interest: Authors declare that they do not have any conflict of interest.

References

  1. Dummit DS, Foote RM. Abstract Algebra, 3rd ed. Wiley; 2003.
     Google Scholar
  2. Jacobson N. Basic Algebra I, 2nd ed. New York: WH Freeman and Company; 1996.
     Google Scholar
  3. Alzubaidy KH, Bennour NM. Matrix representation of group algebras of split metacyclic groups. Int J Math Arch. 2021;12(5):8–10.
     Google Scholar
  4. Alzubaidy KH, Bennour NM. Matrix representations of group algebras of general metacyclic groups. Int J Math Appl. 2021;9(3):33–6.
     Google Scholar