INTRODUCTION:

Functions of square matrices extend scalar functions like powers, exponentials,

trigonometric, and hyperbolic functions to matrices using power series or decompositions.
Matrix powers involve repeated multiplication, while the matrix exponential solves linear
differential systems. Trigonometric and hyperbolic functions model oscillatory and growth
behaviors. General matrix functions apply through series expansions or Jordan forms,
enabling matrices to be treated similarly to numbers in advanced applications.

DEFINITIONS:

Functions of Square Matrices: A function of a square matrix is the extension of a scalar

function to matrices, defined using power series, diagonalization, or Jordan decomposition,
allowing matrices to behave similarly to real or complex numbers under that function.

Matrix Powers:
Matrix powers involve multiplying a square matrix by itself repeatedly, and are defined as A
raised to the power k (written as A^k), which means multiplying A by itself k times. A^0 is
defined as the identity matrix.

Exponential of a Matrix:
The exponential of a matrix is defined by the power series:
e^A = I + A + (A^2)/2! + (A^3)/3! + (A^4)/4! + ... ,
extending the scalar exponential function to matrices.

Trigonometric Functions of Matrices:

Trigonometric functions of matrices, such as sine and cosine, are defined through their Taylor
series expansions by replacing the scalar variable with the matrix, modeling oscillatory
behavior.

Hyperbolic Functions of Matrices:

Hyperbolic functions of matrices, such as sinh and cosh, are defined using their respective
power series expansions, capturing growth and decay phenomena in matrix systems.

General Matrix Functions:

A general matrix function extends any analytic scalar function to matrices, typically through
power series, diagonalization, or Jordan canonical forms, allowing broad functional
operations on matrices.
IMPLEMENTATION (Example Of Types Of Functions Of Square
matrices):

1. Matrix Powers:
When we talk about the power of a matrix A, it’s just repeated multiplication:
Formula:
A^k = A × A × A × ... × A (k times)
Also, A^0 = I (where I is the identity matrix).

Key Properties:

1. A⁰ = I (Identity Matrix)

Just like with numbers, any square matrix to the power of 0 is defined to be the identity
matrix I of the same dimension.

2. A¹ = A

The first power of a matrix is the matrix itself.

3. A² = A × A, A³ = A × A × A, and so on.
4. Matrix Must Be Square

Only square matrices (same number of rows and columns) can be raised to a power.

5. Non-Commutativity

Matrix multiplication is not commutative: AB≠BA. So powers only make sense when

multiplying the same matrix repeatedly.

6. Associativity Holds

Matrix multiplication is associative:

AkAm=Ak+m , and (Ak)m = Akm.

Example 1:
Find the matrix power A² for the matrix
A=|12|
|34|

Step 1: Recall the formula for matrix powers

Matrix powers are calculated by multiplying the matrix by itself. For example, A² = A × A.

Step 2: Multiply the matrix by itself

Given A = | 1 2 |
|34|

To find A², we multiply A by itself:

A² = | 1 2 | × | 1 2 |
|34| |34|

Use standard matrix multiplication:

First row, first column: (1×1 + 2×3) = 1 + 6 = 7

First row, second column: (1×2 + 2×4) = 2 + 8 = 10
Second row, first column: (3×1 + 4×3) = 3 + 12 = 15
Second row, second column: (3×2 + 4×4) = 6 + 16 = 22

So,
A² = | 7 10 |
| 15 22 |

Final Answer:
A² = | 7 10 |
| 15 22 |

Example 2:
If A = [[1, 0, 0],
[1, 0, 1],
[0, 1, 0]],
find A50.
Solution:
The characteristic equation is:

|1-λ 0 0|
| 1 -λ 1 |
| 0 1 -λ | = 0

⇒ (1 - λ)[(-λ)(-λ) - 1] = 0
⇒ (1 - λ)(λ^2 - 1) = 0
⇒ λ = 1, 1, -1

As the eigenvalue 1 is repeated, we use the second method.

Since the matrix is of order 3, we consider:

Φ(A) = A50 = α2* A2 + α1 *A + α0* I

Let λ be an eigenvalue, then:

λ50 = α2 λ2+ α1 λ + α0 ............(1)

Putting λ = 1 and λ = -1 into (1):

1 = α2 + α1 + α0 .....................(2)
1 = α2 - α1 + α0 .....................(3)

Differentiating equation (1) w.r.t. λ:

50 λ49 = 2α2 λ + α1 ................(4)

Putting λ = 1 into (4):

50 = 2α2 + α1 .......................(5)

Solving equations (3), (4), and (5), we get:

α2 = 25, α1 = 0, α0 = -24

Substitute into the expression for A50

A50 = 25A2 - 24I

Now compute A2:

A2 = [[1, 0, 0],
[2, 1, 0],
[1, 0, 1]]

Therefore,
A50 = 25 * [[1, 0, 0],
[2, 1, 0],
[1, 0, 1]]
- 24 * [[1, 0, 0],
[0, 1, 0],
[0, 0, 1]]
 = [[25, 0, 0],
[50, 25, 0],
[25, 0, 25]].

Final Answer:
A50= [[25, 0, 0],
[50, 25, 0],
[25, 0, 25]].

Example 3 :
If A = [[2, 1],
[1, 2]],
find A50 .

Solution:
The characteristic equation of A is:
|2-λ 1|
|1 2-λ|=0

⇒ (2 - λ)2- 1 = 0
⇒ 4 - 4λ + λ2 - 1 = 0
⇒ λ2 - 4λ + 3 = 0
⇒ (λ - 3)(λ - 1) = 0
⇒ λ = 1, 3

(i) For λ = 1, [A - λI] X = 0 gives:

[1 1][x1] = [0]
[1 1][x2] [0]
⇒ x1 + x2 = 0

Putting x2 = -t, x1 = t:
X1 = [-t] = t[-1]
[t] [ 1]
Hence, corresponding to λ = 1, the eigenvector is [1, -1]’.

(ii) For λ = 3, [A - λI]X = 0 gives:

[-1 1][x1] = [0]
[1 -1][x2] [0]
⇒ -x1 + x2 = 0 and x1 = x2

Putting x2 = t, x1 = t:
X2 = [t] = t[1]
[t] [1]
Hence, corresponding to λ = 3, the eigenvector is [1, 1]’.
∴ Modal matrix M = [1 1]
[-1 1]
Now, |M| = 2 ⇒ M⁻¹ = 1/2 * [1 -1]
[1 1]

If A = [a b]
[c d], then A⁻¹ = (1/|A|) * [d -b] [-c a]
Now, D = [1 0]
[0 3], so D⁵⁰ = [1⁵⁰ 0]
[0 3⁵⁰]
∴ A⁵⁰ = M D⁵⁰ M⁻¹ = [1 1][1⁵⁰ 0][1 -1]
[-1 1][0 3⁵⁰][1 1]
= 1/2 * [1 1][1⁵⁰ 0][1 -1]
[-1 1][0 3⁵⁰][1 1]
= 1/2 * [1 + 3⁵⁰ -1 + 3⁵⁰]
[-1 + 3⁵⁰ 1 + 3⁵⁰].

Final Answer:
A50= 1/2 * [1 + 3⁵⁰ -1 + 3⁵⁰]
[-1 + 3⁵⁰ 1 + 3⁵⁰].
2. Exponential of a Matrix:
The matrix exponential is defined by an infinite series, just like the scalar exponential:
Formula:
e^A = I + A + (A²)/2! + (A³)/3! + (A⁴)/4! + ...
(This series goes on forever, but often converges quickly.)

Key Properties of eA :

1. Always defined for any square matrix AAA (unlike, say, A−1A^{-1}A−1, which may not
exist).

2. If A is diagonalizable:

A=PDP−1⇒eA=PeDP−1

And for a diagonal matrix D=diag(d1,d2,...,dn),

eD=diag(ed1,ed2,...,edn)

3. If A and B commute (i.e., AB=BA):

eA+B=eAeB

Example 1:
Find the matrix exponential e^A for the matrix
A=|01|
|00|

Step 1: Recall the formula for e^A

Matrix exponential is defined by:
e^A = I + A + (A²)/2! + (A³)/3! + (A⁴)/4! + ...

Step 2: Find powers of A

First, A² = A × A

Multiply A by itself:
A=|01|
|00|
A×A=
First row: (0×0 + 1×0) = 0, (0×1 + 1×0) = 0
Second row: (0×0 + 0×0) = 0, (0×1 + 0×0) = 0

So,
A² = | 0 0 |
|00|
(A² is the zero matrix.)

Similarly, A³, A⁴, etc. will also be zero matrices because multiplying by zero matrix gives
zero.

Step 3: Substitute into the series

Since A² = 0, A³ = 0, ..., the series becomes:

e^A = I + A

because all higher-order terms are zero.

Step 4: Write I and A

I=|10|
|01|

A=|01|
|00|

Now, add I + A:

e^A = | 1+0 0+1 |

| 0+0 1+0 |

Simplify:
e^A = | 1 1 |
|01|

Final Answer:
e^A = | 1 1 |
| 0 1 |.

Example 2:
Find e^A and 4A if A = [3/2 1/2]
[1/2 3/2].

Solution :

The characteristic equation of A is

|3/2 - λ 1/2|
|1/2 3/2 - λ| = 0

⇒ (3/2 - λ)2 - 1/4 = 0

⇒ 9/4 - 3λ + λ2 - 1/4 = 0
⇒ λ2 - 3λ + 2 = 0
⇒ (λ - 1)(λ - 2) = 0 ⇒ λ = 1, 2.

(i) For λ = 1, [A - λI]X = 0 gives

[1/2 1/2][x1] = [0] By 2R1[1 1][x1] [0]

[1/2 1/2][x2] [0] 2R2[1 1][x2] = [0]

⇒ R2 - R1 [1 0][x1] = [0] ⇒ x1 + x2 = 0
[0 1][x2] [0]

Putting x2 = -t, we get x1 = t.

⇒ X1 = [ t ] = t[ 1 ]
[ -t] [-1]
Hence, the eigenvector is [1, -1]’.
(ii) For λ = 2, [A - λI]X = 0 gives

[-1/2 1/2][x1] = [0] By 2R1[-1 1][x1] = [0]

[1/2 -1/2][x2] [0] 2R2[ 1 -1][x2] [0]

⇒ R2 + R1 [-1 0][x1] = [0] ⇒ -x1 + x2 = 0 ⇒ x1 = x2

[0 1][x2] [0]

Putting x2 = t, x1 = t, we get
⇒ X2 = [ t ] = t[1]
[ t ] [1]
Hence, the eigenvector is [1, 1]’

⇒ M = [1 1]
[-1 1], |M| = 2
⇒ M-1 = adj.M / |M| = 1/2 [1 -1]
[1 1]
Now D = [1 0]
[0 2]

If f(A) = eA, f(D) = eD = [e1 0]

[0 e2]
If f(A) = 4A, f(D) = 4D = [41 0]
[0 42]
⇒ eA = M f(D) M-1 =
= 1/2 [1 1][e 0][1 -1]
[-1 1][0 e2][1 1]

= 1/2 [ e + e2 -e + e2 ]
[ -e + e2 e + e2 ]

Similarly, replacing e by 4 we get,

4A = 1/2 [20 12] = [10 6]
[12 20] [ 6 10]

Final Answer:

eA= 1/2 [ e + e2 -e + e2 ]
[ -e + e2 e + e2 ].
3.Trigonometric Functions of Matrices:
For cosine and sine of a matrix A, the formulas are:

Cosine:
cos(A) = I - (A²)/2! + (A⁴)/4! - (A⁶)/6! + ...

Sine:
sin(A) = A - (A³)/3! + (A⁵)/5! - (A⁷)/7! + ...

(Again, replace the variable x by matrix A in the usual Taylor series.)

Key Properties :

1. Series Definition

Matrix sine and cosine are defined via their Taylor series, just like scalar sine and cosine:

Cosine:
cos(A) = I - (A²)/2! + (A⁴)/4! - (A⁶)/6! + ...

Sine:
sin(A) = A - (A³)/3! + (A⁵)/5! - (A⁷)/7! + ...

2. Only for Square Matrices

These functions are only defined for square matrices, since powers and series expansions
require consistent matrix dimensions.

3. Closely Related to Matrix Exponentials

The matrix exponential can be written in terms of sine and cosine using Euler’s formula:

eiA=cos(A)+isin(A)

4. Useful in Solving Differential Equations

Appears in solutions to second-order systems like:

D2X/DT2=−AX⇒X(t)=cos(tA)X0+sin(tA)X1

5. Computation via Diagonalization or Jordan Form

If A is diagonalizable (or in Jordan form), sine and cosine can be computed easily by
applying the series to the diagonal or Jordan blocks.
Example 1:
Find the cosine of matrix A, cos(A), for the matrix
A=|01|
|00|

Step 1: Recall the formula for cosine of a matrix

The cosine of a matrix A is defined by the Taylor series expansion:
cos(A) = I - (A²)/2! + (A⁴)/4! - (A⁶)/6! + ...

Step 2: Calculate the powers of A

We already know from the Matrix Powers example that:
A=|01|
|00|

And we found that:

A² = | 0 0 |
|00|

Since A² = 0, all higher powers of A (A³, A⁴, etc.) will also be zero matrices. So,
A⁴ = A⁶ = A⁸ = ... = 0.

Step 3: Substitute into the series for cos(A)

Now, substitute the values of A² = 0 into the series:

cos(A) = I - (A²)/2! + (A⁴)/4! - (A⁶)/6! + ...

Since A² = 0, A⁴ = 0, etc., the series becomes:
cos(A) = I - 0 + 0 - 0 + ...

So, cos(A) = I.

Step 4: Write I and the result

I=|10|
|01|
Thus,
cos(A) = | 1 0 |
|01|

Final Answer:
cos(A) = | 1 0 |
| 0 1 |.

Example 2:

If A = [π π/4]
[0 π/2], find cos A.

Solution :

The characteristic equation is

| π - λ π/4 |
| 0 (π/2) - λ | = 0
⇒ (π - λ)(π/2 - λ) = 0 ⇒ λ = π/2, π.

Let Φ(A) = cos A = α₁A + α₀I

Since λ satisfies the above equation, we have
cos λ = α₁λ + α₀

Putting λ = π/2, we get:

cos(π/2) = α₁(π/2) + α₀ ⇒ 0 = α₁(π/2) + α₀ ...(2)
cos(π) = α₁π + α₀ ⇒ -1 = α₁π + α₀ ...(3)

From (2) and (3), we get:

α₁π/2 = -1 ⇒ α₁ = -2/π
⇒ α₀ = -1 + 2 = 1

Putting these values in (1), we get:

cos A = -2/π * [π π/4]
[ 0 π/2] + [1 0]
[0 1]

Final Answer:
cos A = [-1 -1/2]
[0 0].
4.General Matrix Functions

Example:
Find the general matrix function f(A) = e^A for the matrix
A=|01|
|00|

(Note: We’ve already found e^A in earlier examples, so we’re reusing the same matrix for
simplicity.)

Step 1: Recall the formula for general matrix functions

A general matrix function f(A) can be expressed using methods such as power series,
diagonalization, or Jordan canonical forms. For simplicity, let's use the power series
approach, similar to how we calculated e^A.

The power series for a general matrix function f(A) is given by:
f(A) = I + A + (A²)/2! + (A³)/3! + (A⁴)/4! + ...

Step 2: Calculate the powers of A

Given A = | 0 1 |
|00|

We already know that:

A² = | 0 0 |
|00|

Since A² = 0, all higher powers of A (A³, A⁴, etc.) will be zero matrices. So:
A³ = A⁴ = A⁵ = ... = 0.

Step 3: Substitute into the series for f(A)

Substitute the values of A² = 0, A³ = 0, etc., into the series:

f(A) = I + A + (A²)/2! + (A³)/3! + (A⁴)/4! + ...

Since A² = 0, A³ = 0, etc., the series becomes:
f(A) = I + A.

Step 4: Write I and A

I=|10|
|01|

So,
f(A) = | 1 0 | + | 0 1 |
|01| |00|

f(A) = | 1 1 |
|01|

Differences Between types of functions of square matrices:

Aspect Basic Idea Use Case Formula Base

Matrix Powers Repeated Simple computations, Multiplication
multiplication of powers
matrix
Exponential of a Extending exe^xex
Solving differential Power series of
Matrix to matrices using
equations exponential
series
Trigonometric Extending sine, Modeling oscillatory Power series for sine
Functions cosine to matrices behaviors and cosine
Hyperbolic Extending sinh, cosh Modeling growth and Power series for sinh
Functions to matrices decay behaviors and cosh
General Matrix Extending any General functional Depends on the
Functions analytic function to calculations like specific scalar
matrices log⁡(A)\log(A)log(A), function extended
A\sqrt{A}A

Conclusion:
This project explores different types of functions of square matrices, including matrix
powers, exponential, trigonometric, hyperbolic, and general matrix functions. It explains how
scalar functions are extended to matrices through power series, multiplication, and
diagonalization techniques. By understanding these concepts, we can solve complex
problems in systems of equations, differential equations, and mathematical modeling
involving matrices.