0% found this document useful (0 votes)
26 views6 pages

211 PS35 AndalAdrian 230611 083939

To determine a complete set of representatives in M2(Z3): 1) The possible elementary divisors of a matrix in M2(Z3) are x, x+1, x+2, 2x, which yield the rational canonical forms [0] D(0) 02, (-1) D(-1) [2] 12, (-2) D(-2) (1), [0] D(0) [0] respectively. 2) This gives four representatives with the given rational canonical forms, providing a complete set of representatives that satisfies all properties.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
26 views6 pages

211 PS35 AndalAdrian 230611 083939

To determine a complete set of representatives in M2(Z3): 1) The possible elementary divisors of a matrix in M2(Z3) are x, x+1, x+2, 2x, which yield the rational canonical forms [0] D(0) 02, (-1) D(-1) [2] 12, (-2) D(-2) (1), [0] D(0) [0] respectively. 2) This gives four representatives with the given rational canonical forms, providing a complete set of representatives that satisfies all properties.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

ADRIAN J. ANDAL.

Problem Set 3.5

MATH 21.

1. Let CMCF). We say that


is a
complete set ofsimilarity
representatives in MCF) if the

following are satisfied.

i) every At MCF) is similar to some SE,


ii) if R, T E sit. RFT, then R is not similar to, that is, I contains exactly one

class in MCF,
representative from each similarity

Determine a
complete set of representatives
similarity in Me (2).

(Ez 30,1,23 is the field under addition and multiplication modulo 3)


=

solution:
By Rational Canonical form theorem, two matrices in Mels) are similar iff. They have the same rational canonical

form and ifthey have the same elementary divisors.

By the properties of minimal polynomial, the lowest and highest possible degree of the minimal polynomial of a matrixin Mc(z)
must be 1 and 2. Hence if AGMe(s), then Amust satisfy the equation ax+ bx+c =

0 for some a, b, c t R,

Note that if a b
=

c
=

0, then the
=

degree of polynomial would be O, thus, it is not possible since the lowest degree must be 1.

Similar argument shows when a b


=

0
=

and c =
1 and c 2.
=

·
If a
=

c 0
=

and b 1, then Asatisfies


=
X o.=

Since malls and so is irreducible in Ig[C],then mac =x.. Hence,


the only possible elementary divisor is. So, we can add
copies of ce so that the sum of all the degrees is 2,which is

the size of A.

Thus, the possible elementary divisor:x,which has the RCF given as [0]8 [0] 0, =

For the
remaining possibilities, we follow the same
argument.
·
If a
0,c
= =
1 and b 1 =

A satisfies
=>

XH 0
=

(xt is irreducible in 2,(X))


MA(x)
=>

x
=
+

->
possible elementary divisors:XH1, x

=>
Rational canonical form:(-1) D(-1) (27*[2) 212
=
=
since -1 2 in
=

s.

·
If a 0,b 1,c 2
= = =

=>
Asatisfies x +2 0
=

(xt2 is irreducible in
Zc,[X])
=>
Mp(X) x 2
= +

possible elementary
=>
divisors:X2, X+2

rational canonical form:(-2]


->
D(-2) =
(1) D(1) 12 =

·
If a 0,b 2,c
=
=
0
=

=>>
Asatisfies 2x 0
=

=>
> ma(x)/2 x => ma(x) x
=

possible elementary divisors:X,


=>
X

=>
Rational Canonical form:[O] D(OJ O2 =
·
If a 0,b 2,c
=
=
1
=

=>
Note that the equation 2xH does not have any roots in , and hence, 2xH is irreducible
that
Also, note 2x+1 is the lowest degree but it is not monic. Therefore if AGMc (E), then ma(x) +2x+1.

·
If a 0,b 2,c 2= = =

Asatisfies the
=>

equation 2x+2 0
=

(2x 2 +
2(x
=

1)7
+

MA(x) x +1
=> =

I
possible elementary divisors:X+1, x+ 1

Rational Canonical form:


=>
(-17*[-1]=[2]D(2) t
=

· If a =

1,b 0,c0 =
=

(Xx,x)
=

satisfies the equation


A
x 0
A
=

=>

ma(x)/x2 max) X. Now, if max) x, then On


=

x or
=
=
=
=>

elementary divisors:X2
I
possible
->
Rational Canonical form if ma(X)= x:(98]
·
If a 2,b 0, =0
= =

=>
Asatisfies 2x2 0 =

(2x 2(x2) =

ma(X)/2x2
2
->
=> ma(x) x =

or X

I note that if ma(x) x, then =


A O2, and if ma(x) x 2, then
= =

A
(98]
=

· If a 1,b 0,c 1
= =
=

A satisfies the
=>

equation x2+1 0
=

MA(x) x+
=
=>

(9]
elementary divisors:X2H
I
possible
Rational canonical form ifma(x)
=>

x2+1:(907
=
=

· If a 1,b 1, =0
= =

A satisfies the equation x+x 0


=

=>

2
X
MA(x) x +
=>
=

elementary divisors:xx
I
possible
Rational Canonical form:
=>

[ 99] 9 =

· If a 1,b 0,c
= =

2
=

=>
Asatisfies the equation x+ 2 0 =
(x+2 (x 2)(x +)
= +

ma(x) (x 2),(x 1)
=> =
+ +

or x+ 2

possible elementary
=>
divisor:x 2.

1, (i0]
Rational canonical form:
=>

Note that If MA(X) x+2, =


A 12
=

X+1, A
=
212
=
a
Ifa 2,b 0,c 1 =
=
=

=>
Asatisfies the equation 2x2+1 0
=

(2x+)
=

2(x+ 1)(x1))
+

mA(x) (x 2),(x 1)
=> =
+ +

Note that If MA(X) x+2, =


A 12
=

X+1, A
=
212
=

· If a 2,b 0,c 2
= = =

=>
Asatisfies the equation [x+C 0
= (2x2+2 2(x2 1)
=
+

2
mq(X) + 1
=>

x
=

Now, If ma(X) x2
=

1, then
+

A =

[ii)
·
Ha 2,b 1,c = =
1
=

A satisfies 2x2 +xH 0


=>
=

(2x4+x+1 2(x2 2x 2)) =


+
+

mq(x) x2
=>
=
2x 2
+
+

possible elementary divisor:x2+2x + 2


=>

[ii) ii)
Rational
=>

Canonical form:

·
If a 2,b 1, =2
=
=

=>
Asatisfies 24+x+ 2 =

0 (2x4+ x 2 2(x+12)
+ =

MA(X) x+2x + 1
=> =

=>

possible elementary divisor;x2+2x+

(P-7=(,]
->
Rational canonical form;

·
If a =

2,b 2,c = =

A
=>
satisfies 2x+ 2x +1 (2x 2x+ 2(x+x
+ =
2))
+

mA(x)
=>
xx 2
=
+
+

possible elementary
=>
divisor:x 2+x+2

(P?]=(i el
=> Rational canonical form:

·
If a 2,b 2, 2
= =
=

1)
2
=> Asatisfies 2x2+2x +2 0
=

(2x +2x +2 2) x
=

x
+
+

MA(x) x2+ x 1
=> = +

possible elementary
=> divisor:x27xH

(iii) (i ij
=>
Rational Canonical form:
=

*
if a
1,b 1,c 1
=
= =

A satisfies
=>
x2 + x+ 1 0
=

+
ma(X) x x 1
=> =
+

possible elementary divisor: x2 + x+1


=>

[P]=(i2]
Rational canonical fom:
->
If a
1,b 1,2 2
= =
=

If a 1,b2,c 1
=
=
=

A satisfies
-> x2 + x +
2 0 =

A
=>satisfies x+ 2x+1 0
=

mq(x) x2+ x
=>
2 MALX x+2x + 1
=>
=
+ =

if ma(x) x x
=> = +
2
+
then, A
(ic]
=
->
if MA(x) x2 2x+1, then =
+

A =

1,]
if a 1,b 2,c 2
=
=
=
·
if a 2,b 2,c
= =
0
=

A satisfies
-
x+ 2x 2
+

0
=

A
=>satisfies 2x + x 0 =
(2x+x 2(x2+x))
=

2
MA(x) x 2x 2 MA(X) x, x+1 x2X
=>

or
=
+ =
+ =>

If MACA) x
=> =
2x+2
+
then, A =

(9,7 If
=>

MA(x) X, then A
=
=
02
A 2 12
MACX) =
XH thew =

(P2]
ma(x) x2 + x
=
then A =

·
If a
1,b 2c 0
=

= =

A satisfies
=>
x+ 2x 0
=

(x2 + 2x x(x 2))


=
+

=>
mACX) X
=
or x+2 or x+2x

if mq(X) x, then
=
A 02
=

MACX) x+2, their A I2


=
=

If
->

ma(x) x
= +
2x

possible
=>
elementary divisor:x+ 2x
Rational canonical form:
[ 9.2]:[ii]
=>

Therefore, a complete of
set
similarity representatives in Me (Is] are:

9a,252, In, (98], [98], (ii),(i:], (9:], [92]. (92]

(92] (;;]3
2)Let F AR or D, and
IV with inner product C, 7. Suppose I is positive integer and S
G.,...,VICV.
=
=

Let A
[9ij] EM, CE <Vi,Vi>
sit.
Gij:= i,j. for all
=

is
show that if A non-singular, then S is linearly independent over Fi

CA here is an
arbitrary vector space with an inner
product (h)

Proof:
If Ais need to show that S is linearly independent F.
non-singular, we over

To answer this, we will use


contrapositive logic wherein "if p,
then g"is equivalentto "if ug, then up,"

Now, suppose that S is F, then, it


linearly dependent over can be assumed that

Up GVitGeVet
=
...

OpYk- for 2,2, ...,


GEF.
Consecutively,
[V, Vi7 <C,V, GzVzt 8Vk VinL
=

...
-
1,

,V, Vi +[Va, Vin) +...Vis-,Vish


-
(since the inner
product is
conjugate linear in the first component.
CV, m) +CV2,Vist... GU, V
=

(since , ..., are scalars, its conjugate is still the same

MatrixA,
Now, consider

Caniai)- INine-wi]-frn viewtaste, I


SV, V, SV, V2... [V, Vi
A
... "

rest ...

cree

Let's apply the RREF.


Consider the row operation Ri -RK -
(c, R, tc Ret... GRk-) by first taking the last element in the
final row, then,

I wereis e
SV, V, SV, V2... [V, Vi

cnn
A
=

....

taste,rest ...

Gum, -(a.r, ntaSin,rest... cure

IViv re-Sr I
SV, V, SV, V2... [V, Vi

e
-

...

Now, for any other element, which is of the form [V,vi) will be
where Ij<k-1, the computation

<V, Vi -[35V,;D +2[Ve,Vj7 +... (,V,-,Vj] +


=

(Vx,Vi) [[,V,,Vj2 [Va,Vj7+ [Cr-V, V57]


- +
...+

[Vk, Vj7 [[CV, tGVst... CV, V5L] (by Linearity)


=
-
+

=[Vi -(2,r, +GVz .. (,Y-),Vi7


+
+
.

-[V-V, Vj7 (Since Vi C, r, +CaUzt


=

...
GnVK 1) -

70,Vj7
=

=0. (from remark that states (2,07 0 50,n))


=
=
After the operation to all entries the final row, have,
applying row in we now

I I
SV, V,7 <r,,Ve] ...
SV, Vin

(V,r> <r2, Vah.r


e
A
=

This means that there


are n-1
linearly independent rows of A. Thus, rank (A)=n-1. Recall from proposition 5.3 that a matrix
A is

non-singular iff rank (A) n. Since


=

rank CA) In,then A is singular,


Therefore, if S is linearly dependent then A
is
singular, Contra positively, Ais non-singular LS is linearly independent F
F,
over
over

You might also like