August 2017
Contents
Introduction
Prerequisites:
I Differential equations (i.e., Math knowledge).
I Electric circuits I (i.e., basic Physics knowledge - dynamic systems).
Signals, systems and control syllabus @ UN (undergraduate):
I SyS I: time-domain descriptions of signals and systems, fundamental transforms,
basic applications.
I SyS II: advanced filter theory and realization, linear systems and feedback theory.
I Control: linear control theory (classical), state-space control design.
x2 4 = 0 x2 = 4
j 2 = 1 j 3 = j 2 j = j j 4 = j 2 j 2 = 1 and so forth
ja and jb
Complex number
Sum of a real number and an imaginary number
a + jb and b + ja
For this course we will adopt the handwritten material convention for complex
numbers (in both handwriting and computer typing).
Complex plane
A = a + jb A = a + jb
jb jb
a Re a Re
jIm
Magnitude of a complex number
The length cannot be negative A = (rA )
Example 0.1
Express the complex number 3 + j2 in polar form and plot it on the complex plane
p q
r= a2 + b 2 = (3)2 + (2)2 = 3.6
b 2
= tan1 ( ) = tan1 ( ) = 33.69
a 3
Thus the polar representation is:
(r ) = (3.6 33.69 )
Is this correct?
Angles-modulo
jIm
2 = 45 Re
Then,1 = 315 and 2 = 45 are
the same angle
The concept of modulo helps to
A
make the discussion precise
Figure: Angles-modulo.
Angles-modulo
Definition
Two angles 1 and 2 are considered the same angle if they differ by 360 or its
integer multiple, or, equivalently, their difference can be divided by 360 wholly:
1 = 2 (mod 360 )
(5)
1 2 = k 360 for some integer k
Exercise 0.1
Are the angles +56 and 304 the same angle?
1 2 = 56 (304 ) = 360 = (1)(360 )
More generally:
56 = 416 = 776 = = 304 = 664 (mod 360 )
sin ( ) = sin
cos ( ) = cos
we get the negative-angle or conjugate form:
e j = cos j sin
and after combining this result with the Eulers formula, we obtain the exponential
expression for the sine and cosine functions:
e j + e j e j e j
cos = sin =
2 2j
S. A. Dorado-Rojas Mathematical Foundations
Introduction Complex numbers Complex matrices Series
A = a + jb = (r ) = re j
|{z}
| {z } | {z }
rectangular form polar form exponential representation
Note that the exponential representation is closely related to the polar form:
r = a2 + b 2
Im{A}
= tan1 ba = tan1 Re A
{ }
r cos = a r sin = b
As far as possible, we will use:
I Polar representation: with engineering notation (angle in degrees).
I Exponential representation: with mathematical notation (angle in radians as an
irrational number: fraction of )
S. A. Dorado-Rojas Mathematical Foundations
Introduction Complex numbers Complex matrices Series
Exercise 0.2
Exercise 0.2
Express -3 in polar form
3 = 3e j0 =? (30 )
The last expression is not properly expressed in polar form because we require r 0!
Thereby, a correct polar expression for this negative number is:
3 = (3180 ) = 3e j180 = 3e j
Complex conjugate
Definition
Let A = a + jb be a complex number. Then
jIm
A is known as the complex conjugate of A:
A
A = a jb (6) jb
2. A A = j2Im A
A+A
3. A = 2
4.
AA = a2 + b 2 = A2 = |A|2
5. (A B) = A B
6. (AB) = A B
Quotient:
A AB AB
= =
B B B B2
This operation is simplified in exponential form:
A Ae j A
= j
= e j( )
B Be B
1 1
An = An e jn = = n
An e jn A (cos n + j sin n)
Matrix
An nm (n by m) matrix is a rectangular array of objects arranged in n rows and m
columns. We will denote matrices in boldface:
a11 a12 . . . a1m
a21 a11 . . . a2m
Anm = . .. = [aij ]
.. ... ... .
an1 an2 . . . anm
where aij is known as the i, j element: the object located in the row i and column j
place of a matrix.
If the elements of a n m matrix are real numbers (real-valued matrix), then each row
can be thought as a vector in Rm and each column as a vector in Rn :
r1,1xm
r2,1xm
Anm = c1,nx1 c2,nx1 . . . cm,nx1 =
...
rn,1xm
Equality
Two matrices A = [aij ] and B = [bij ] are equal if:
I they have the same number of rows
Addition of matrices
If A = [aij ] and B = bij are both n m matrices, then their sum is defined to be the
n m matrix:
Multiplication by a scalar
If A = [aij ] and c C, then
cA = [c aij ]
The operations of matrix addition and multiplication by a scalar enable one to write
any complex matrix as the sum of two real matrices, the imaginary part being
multiplied by the imaginary operator:
Matrix multiplication
Let A = [aij ] be n k and B = [bij ] be k m. Then the product AB is the n m
matrix whose i, j element is
k
ais bsj = ai1 b1j + ai2 b2j + + aik bkj
s=1
I AB 6= BA
1 0 2 6 2 6
=
2 4 2 3 8 0
but
2 6 1 0 10 24
=
2 3 2 4 7 12
AT = [aji ]
For example:
1 6 3 4
A24 =
0 12 5
1 0
6
(A24 )T = AT
42 =
3 12
4 5
S. A. Dorado-Rojas Mathematical Foundations
Introduction Complex numbers Complex matrices Series
T
I For any matrix A, AT =A
I (cA)T = cAT
Inverse of a matrix
Inverse of a matrix
Let A be a n n matrix. An n n matrix B is the inverse of A if
AB = BA = In
I If A and B are nonsingular, then so is AB. Further, the inverse of a product is the
product of the inverses in the reverse order:
(AB)1 = B1 A1
I If A is nonsingular, so is A1 . Furthermore, the inverse of the inverse is the matrix
itself:
1
A1 =A
Generalized inverse
Generalized inverse
When a matrix is square and it has an inverse (i.e., it is nonsingular), then this inverse
is its unique generalized inverse.
Matrix division
In real numbers, dividing by a nonzero number is equivalent to multiply by the
reciprocal of the divisor:
a 1
= a
b b
In matrices, we may define a similar operation.
x=A\B or x=mldivide(A,B)
S. A. Dorado-Rojas Mathematical Foundations
Introduction Complex numbers Complex matrices Series
Matrix determinant
Let M be the vector space comprising square matrices. The determinant is a function
or mapping from C to the complex number set C that assigns to each matrix A an
unique complex number det A:
M C
Ann det A = |A|
Cofactor of an element
Let A be n n. The i, j cofactor cij of A0 s i, j element is:
cij = (1)i+j det (Mij )
where:
I Mij is the matrix formed my removing row i and column j from A, known as the
minor ij
where ckm represents the cofactor associated with the element akm
det B = det A
det B = det A
I If A is scaled by
det (A) = n det (A)
I If B is formed from A by adding times one row (or column) to another row (or
column), then:
det B = det A
I det(A) in MATLAB
I Det[A] in Mathematica
Ax = x
Ax = x
for nonzero vectors x. We can rearrange this equation as:
Ax x = 0
(A In ) x = 0
For this system to have a nontrivial solution x 6= 0, the matrix A In must be
singular, which can be expressed as a single value expression as:
det (A In ) = 0
This last expression corresponds to a polynomial equation on of degree n known
as characteristic equation.
S. A. Dorado-Rojas Mathematical Foundations
Introduction Complex numbers Complex matrices Series
If the n eigenvalues of A are different and non repeated, then we call them
nondegenerate eigenvalues. If the multiplicity of an eigenvalue is greater than one
(i.e., it is a repeated root of the characteristic polynomial), we call it a degenerate
eigenvalue.
If A is real-valued and 1 = a + jb is an eigenvalue, then 2 = a jb = 1 is also an
eigenvalue of A.
The product of the n eigenvalues of A equals the determinant:
n
det A = i = 1 2 . . . n
i=1
The trace of the matrix A is defined as the sum of the entries on the main diagonal:
Hermitian operator
Hermitian operator
The Hermitian or adjoint operator is defined as taking the conjugate transpose of
a matrix:
AH (A )T
Important: the MATLAB transpose command automatically takes complex
conjugates: A = AH .
For real matrices, the Hermitian is the same as the transpose operation. Hence, all
known properties of transposing are applicable:
H
I AH =A
I (A + B)H = AH + BH
I (A)H = AH for real
I (AB)H = BH AH
uH v = u1 v1 + u2 v2 + + un vn
In general, uH v 6= vH u = v1 u1 + v2 u2 + + vn un but
uH v = v H u
A zero inner product means that the complex vectors are orthogonal.
Symmetric matrix
A symmetric matrix is a real square matrix whose transpose equals itself
A = AT
A symmetric matrix A has the properties:
I All eigenvalues of A are real.
I All eigenvectors associated to nondegenerate eigenvalues conform an orthogonal
set.
Orthogonal matrix
An orthogonal matrix is a real square matrix whose transpose equals its inverse
AT = A1 , and, equivalently
AT A = AAT = In
The determinant of an orthogonal matrix can only be det A = 1 and its eigenvalues
can be only have | | = 1.
Hermitian matrices
Hermitian matrices
Let Ann be a complex matrix. A is a Hermitian matrix if
AH = A
Unitary matrices
Unitary matrices
A unitary matrix Un is a (complex) square matrix that has orthonormal columns.
Therefore:
U H U = In
A unitary matrix is characterized by the fact that all its eigenvalues U have magnitude
equal to one
|U | = 1
The concept of unitary matrix for complex matrices is the extension of that of
orthogonal matrix (AT A=In ) for real matrices.
Normal matrices
Normal matrices
A normal matrix Nnn is a matrix that conmutes with its Hermitian:
NH N = NNH
N = 1
AxR, = R xR
Theorem
If A is a symmetric real matrix, then the left and right eigenvectors are simply each
others transpose:
xL = xT R
Theorem
If A is Hermitian (self-adjoint), then the left and right eigenvectors are adjoint matrices:
xL = xH
R
Sequences
Limit of a sequence
A sequence an has the limit L and we
Sequences write
A sequence an is a list of numbers
written in a definite order lim an = L
n
If limn an exists, we say that the sequence converges. Otherwise, we say that the
sequence diverges.
Series
Series
If we add the terms of an infinite sequence an we get an expression called an infinite
series
an = a1 + a2 + + an + . . .
n=1
Series
Partial sums
Given a series
n=1 an we denote its nth partial sum as
n
sn = ai = a1 + a2 + + an
i=1
If the sequence is convergent, and limn sn = s exists as a real number, then the series
is called convergent and we write
n
lim sn = s
ai = s
i=1
I Maclaurin series
f (n) (0) n
f (x) = n! x
n=0
a
ar n1 = 1 r |r | < 1
n=0