MKT3421 Signals and Systems

Fourier Analysis: The Fourier Series

Prof. Aydın Yeşildirek

Mechatronics Department

Signals and Systems

Signal representations
● Every continuous function in the function space can be represented as a linear combination of some basis
functions

𝑥 𝑡 = ෍ 𝑋𝑘 𝜑𝑘 (𝑡)

● For the periodical signals such a set of basis functions composed of

− Fourier basis (eigen functions)

𝜑𝑘 𝑡 = 𝑒 𝑗𝑘Ω𝑜 𝑡

● We may even consider the polynomials

𝜑𝑘 𝑡 = 𝑡 𝑘

Then the Taylor series expansion may be considered as another example

Prof. A. Yesildirek Fourier Series 2

 Signals and Systems

Fourier Basis
● For the periodical signals the Fourier basis
𝜑𝑘 𝑡 = 𝑒 𝑗𝑘Ω𝑜 𝑡
− Frequency Ω𝑜 = 2𝜋𝑓𝑜 = 2𝜋/𝑇𝑜
● 𝜑𝑘 are an ortho-normal set

1 𝑡𝑜 +𝑇𝑜 1 𝑡𝑜 +𝑇𝑜 𝑗(𝑘−𝑙)Ω 𝑡

න 𝜑𝑘 𝑡 𝜑𝑙∗ 𝑡 𝑑𝑡 = න 𝑒 𝑜 𝑑𝑡
𝑇𝑜 𝑡𝑜 𝑇𝑜 𝑡𝑜

𝑒 𝑗 𝑘−𝑙 2𝜋 − 1
= 𝑒𝑗 𝑘−𝑙 2𝜋
𝑗 𝑘 − 𝑙 2𝜋

0 𝑘≠𝑙
=ቊ
1 𝑘=𝑙

Prof. A. Yesildirek Fourier Series 3

Signals and Systems

Example
● Let

● Eulers identity

● Rearrange

● or

Prof. A. Yesildirek Fourier Series 4

 Signals and Systems

Example
● Let
𝑥 𝑡 = 2 + 4 cos 4𝑡 + 6 sin 6𝑡 + 8 cos 10𝑡

● Find 𝑋2 , 𝑋−3 , 𝑋4

Periodic? If so, the fundamental frequency?

2𝜋 𝜋 2𝜋 𝜋 2𝜋 𝜋
𝑇1 = = , 𝑇2 = = , 𝑇3 = =
4 2 6 3 10 5
● Periodicity
𝑚𝑇1 = 𝑘𝑇2 = 𝑙𝑇3

● Integers, for 𝑚 = 2, 𝑘 = 3, 𝑙 = 5 → 𝑇𝑜 = 𝑚𝑇1 = 𝑘𝑇2 = 𝑙𝑇3 = 𝜋

Ωo = 2

Prof. A. Yesildirek Fourier Series 5

Signals and Systems

3
𝑥 𝑡 = 2 + 2 𝑒 𝑗4𝑡 + 𝑒 −𝑗4𝑡 + (𝑒 𝑗6𝑡 − 𝑒 −𝑗6𝑡 ) + 4(𝑒 𝑗10𝑡 + 𝑒 −𝑗10𝑡 )
𝑗

𝑥 𝑡 = 2 + 2 𝑒 𝑗2𝛺𝑜 𝑡 + 𝑒 −𝑗2𝛺𝑜 𝑡 − 3𝑗(𝑒 𝑗3𝛺𝑜 𝑡 − 𝑒 −𝑗3𝛺𝑜 𝑡 ) + 4(𝑒 𝑗5𝛺𝑜 𝑡 + 𝑒 −𝑗5𝛺𝑜 𝑡 )

𝑋2 = 2,
𝑋−3 = 3𝑗,
𝑋4 = 0

Prof. A. Yesildirek Fourier Series 6

 Signals and Systems

Fourier Series Definition

● Any periodic signal 𝑥 𝑡 with the period 𝑇𝑜 or frequency Ω𝑜 = 2𝜋/𝑇𝑜 may be expressed by Fourier series

𝑥 𝑡 = ෍ 𝑋𝑘 𝑒 𝑗Ω𝑜 𝑘𝑡
𝑘=−∞

With the complex Fourier Series coefficients

1 𝑡𝑜+𝑇𝑜
𝑋𝑘 = න 𝑥(𝑡)𝑒 −𝑗𝑘Ω𝑜𝑡 𝑑𝑡
𝑇𝑜 𝑡𝑜

Prof. A. Yesildirek Fourier Series 7

Signals and Systems

Example
𝑥(𝑡)
● Fourier series representation of
A

− Ω𝑜 =?

𝑇𝑜 /2
1 𝜏 𝜏 𝜏 𝜏
𝑋𝑘 = න 𝑥(𝑡)𝑒 −𝑗𝑘Ω𝑜 𝑡 𝑑𝑡 − 𝑇𝑜 − − 𝑇𝑜 −
𝑇𝑜 2 2 2 2
−𝑇𝑜 /2

𝜏/2 𝜏/2
1 𝐴 𝑗𝑘2𝜋𝑡
න 𝐴𝑒 −𝑗𝑘2𝜋𝑡/𝑇𝑜 𝑑𝑡 =
− ቚ
𝑋𝑘 = 𝑒 𝑇𝑜 𝑡=−𝜏
𝑇𝑜 −𝑗2𝜋𝑘 2
−𝜏/2 𝑗𝑘𝜋𝜏 𝑗𝑘𝜋𝜏
−
𝐴 𝑒 𝑇𝑜 −𝑒 𝑇𝑜 sin 𝑥
𝐴 → 𝑠𝑖𝑛𝑐 𝑥 =
𝑋𝑘 = 𝑒𝑗𝑘𝜋𝜏/𝑇𝑜 − 𝑒 −𝑗𝑘𝜋𝜏/𝑇𝑜 𝜋𝑘 𝑗2 𝑥
𝑗2𝜋𝑘

𝜏/𝑇𝑜 1
𝑋𝑘 = 𝐴 sin 𝑘𝜋𝜏/𝑇𝑜
𝜏/𝑇𝑜 𝑘𝜋

𝜏 𝜏
𝑋𝑘 ≝ 𝐴 𝑠𝑖𝑛𝑐(𝑘 )
𝑇𝑜 𝑇𝑜

Prof. A. Yesildirek Fourier Series 8

 Signals and Systems

LTI Response to Periodic Inputs

● Recall if input to an LTI system is an eigenfunction 𝑥 𝑡 = 𝑒 𝑗Ω𝑖 𝑡 , its output is
𝑦 𝑡 = 𝑥 𝑡 𝐻(𝑗𝛺𝑖 )

● The periodic signals in Fourier series representation are the infinite sum of eigen functions 𝑒 𝑗Ω𝑜 𝑘𝑡 , i.e.,

𝑥 𝑡 = ෍ 𝑋𝑘 𝑒 𝑗Ω𝑜 𝑘𝑡
𝑘=−∞

● Using the LTI eigenfunction property, response to periodical inputs represented by fourier series

∞ ∞

𝑦 𝑡 = ෍ 𝑋𝑘 𝐻(𝑗Ω𝑜 𝑘) 𝑒 𝑗Ω𝑜 𝑘𝑡 = ෍ 𝑌𝑘 𝑒 𝑗Ω𝑜 𝑘𝑡

𝑘=−∞ 𝑘=−∞

𝑌𝑘 = 𝑋𝑘 𝐻(𝑗Ω𝑜 𝑘)

Prof. A. Yesildirek Fourier Series 9

Signals and Systems

● Find 𝐻 𝑗Ω of the averager

1 𝑡
𝑦 𝑡 = න 𝑥 𝑎 𝑑𝑎
𝑇 𝑡−𝑇

using

a) Eigenfunction property, if it is applicable,

b) Impulse response

a) LTI? Yes

b) h(t) or H(s)?

Prof. A. Yesildirek Fourier Series 10

 Signals and Systems

b) impulse response

1 𝑡 1 𝑡 1 𝑡−𝑇
𝑦 𝑡 = න 𝑥 𝑎 𝑑𝑎 = න 𝜹 𝑎 𝑑𝑎 − න 𝜹 𝑎 𝑑𝑎
𝑇 𝑡−𝑇 𝑇 −∞ 𝑇 −∞

𝑢 𝑡 −𝑢 𝑡−𝑇
ℎ 𝑡 =
𝑇

1 1 − 𝑒 −𝑠𝑇
𝐻 𝑠 =
𝑇 𝑠

Prof. A. Yesildirek Fourier Series 11

Signals and Systems

● If the averager is driven by a periodic input with Fourier series coefficients

𝑋𝑘

● Then, the Fourier series coefficients of the output

𝑌𝑘 = 𝑋𝑘 𝐻(𝑗Ω𝑜 𝑘)

1 − 𝑒 −𝑗Ω𝑜 𝑘𝑇
𝑌𝑘 = 𝑋𝑘
𝑗Ω𝑜 𝑘𝑇

Prof. A. Yesildirek Fourier Series 12

 Signals and Systems

General Fourier Series

● In general, the Fourier coefficients are complex
𝑋𝑛 = 𝑎𝑛 + 𝑗𝑏𝑛 = 𝑋𝑛 𝑒 𝑗𝜃𝑛

− Amplitude Line Spectrum as a function of 𝑛Ω𝑜

|𝑋𝑛 | = 𝑎𝑛2 + 𝑏𝑛2

− Phase Line Spectrum as a function of 𝑛Ω𝑜

𝑏𝑛
𝜃𝑛 = atan
𝑎𝑛

− Power Line Spectrum as a function of 𝑛Ω𝑜

𝑋𝑛 2 = 𝑎𝑛2 + 𝑏𝑛2

Prof. A. Yesildirek Fourier Series 13

Signals and Systems

Example: Fourier series

1 1 0 1 1
𝑋𝑛 = න 𝑥 𝑡 𝑒 −𝑗𝑛Ω𝑜𝑡 𝑑𝑡 = න −𝐾 𝑒 −𝑗𝑛𝜋 𝑡 𝑑𝑡 + න 𝐾𝑒 −𝑗𝑛𝜋 𝑡 𝑑𝑡
𝑇𝑜 <𝑇𝑜> 2 −1 2 0

1 𝐾 0,0 1 𝐾
= ⋅ ⋅ 𝑒 −𝑗𝑛𝜋 𝑡 ቚ = ⋅ ⋅ 1 + 1 − 𝑒 −𝑗𝑛𝜋 + 𝑒 𝑗𝑛𝜋
2 𝑗𝑛𝜋 𝑡=−1,1 2 𝑗𝑛𝜋

𝐾 𝐾 𝑛
= ⋅ 1 − cos 𝑛𝜋 = ⋅ 1 − −1 , 𝑛 = 0, ±1, ±2, …
𝑗𝑛𝜋 𝑗𝑛𝜋
𝑇𝑜 = 2
2𝜋
Ω𝑜 = =𝜋 2𝐾 −𝑗𝑠𝑖𝑔𝑛(𝑛)𝜋/2
𝑇𝑜
𝑋𝑛 = ቐ 𝑛𝜋 𝑒 , 𝑛 = 𝑜𝑑𝑑
0, 𝑛 = 𝑒𝑣𝑒𝑛

Prof. A. Yesildirek Fourier Series 14

 Signals and Systems

2𝐾 −𝑗𝑠𝑖𝑔𝑛(𝑛)𝜋/2
𝑋𝑛 = ቐ 𝑛𝜋 𝑒 , 𝑛 = 𝑜𝑑𝑑
0, 𝑛 = 𝑒𝑣𝑒𝑛

● 𝑋𝑛 is called the nth harmonic of the signal at frequency of 𝑛Ω𝑜

Prof. A. Yesildirek Fourier Series 15

Signals and Systems

Where are we now?

● We have defined the Fourier series representation of periodic signals
− Composed of complex coefficients multiplied by fourier basis at the fundamental frequency and its
harmonics
− A sum of infinite series may be required though
● Looked at line spectrum
− We had a glimpse of the frequency content of the signal with complex values
− Visualized in terms of its amplitude and phase
− Observed
• Even symmetry of the amplitude line spectrum
• Odd symmetry of the phase line spectrum

● Now let’s have a look at some special cases and properties of the Fourier series

Prof. A. Yesildirek Fourier Series 16

 Signals and Systems

Real-Valued Signals’ Fourier Series

● If we have a real-valued periodical signal, i.e., 𝑥 ∗ 𝑡 = 𝑥 𝑡 = 𝑎 𝑡 + 𝑗0

∞ ∞ ∞

𝑥 𝑡 = ෍ 𝑐𝑛 𝑒 𝑗𝑛Ω𝑜 𝑡 = ෍ 𝑐𝑛∗ 𝑒 −𝑗𝑛Ω𝑜 𝑡 = 𝑐𝑜 + 2 ෍ 𝑐𝑛 cos(𝑛Ω𝑜 𝑡 + 𝜃𝑛 ) + 𝑗 ∙ 0

𝑛=−∞ 𝑛=−∞ 𝑛=1

nth harmonic
● Real-valued signal’s Fourier series representation may be expressed with real coefficients

𝑥 𝑡 = 𝑎𝑜 + 2 ෍ 𝑎𝑛 cos(𝑛Ω𝑜 𝑡) + 𝑏𝑛 sin(𝑛Ω𝑜 𝑡)
𝑛=1

Prof. A. Yesildirek Fourier Series 17

Signals and Systems

● Fourier series coefficients become

1
𝑎𝑛 = න 𝑥 𝑡 cos 𝑛Ω𝑜 𝑡 𝑑𝑡 = 𝑐𝑛 cos 𝜃𝑛 , 𝑛 = 0, 1, …
𝑇𝑜 <𝑇𝑜 >

1
𝑏𝑛 = න 𝑥 𝑡 sin 𝑛Ω𝑜 𝑡 𝑑𝑡 = − 𝑐𝑛 sin 𝜃𝑛 , 𝑛 = 1, 2, …
𝑇𝑜 <𝑇𝑜 >

With

𝑐𝑛 = 𝑎𝑛2 + 𝑏𝑛2

𝑏𝑛
𝜃𝑛 = − atan
𝑎𝑛

Prof. A. Yesildirek Fourier Series 18

 Signals and Systems

The same example treated as real-valued signal

1 1 0 1 1
𝑎𝑛 = න 𝑥 𝑡 cos 𝑛Ω𝑜 𝑡 𝑑𝑡 = න −𝐾 cos 𝑛𝜋𝑡 𝑑𝑡 + න 𝐾 cos 𝑛𝜋𝑡 𝑑𝑡
𝑇𝑜 <𝑇𝑜> 2 −1 2 0

1 0 1 0
= න −𝐾 cos 𝑛𝜋𝑡 𝑑𝑡 + න −𝐾 cos 𝑛𝜋𝑡 𝑑𝑡
2 −1 2 1

𝐾 1 0,0
=− ⋅ ⋅ sin 𝑛𝜋𝑡 ቚ = 0, 𝑛 = 0, 1, 2, …
2 𝑛𝜋 𝑡=−1,1

1 1 0 1 1
𝑏𝑛 = න 𝑥 𝑡 sin 𝑛Ω𝑜 𝑡 𝑑𝑡 = න −𝐾 sin 𝑛𝜋𝑡 𝑑𝑡 + න 𝐾 sin 𝑛𝜋𝑡 𝑑𝑡
𝑇𝑜 <𝑇𝑜> 2 −1 2 0

𝐾 1 0,0 𝐾 1
= ⋅ ⋅ cos 𝑛𝜋𝑡 ቚ = ⋅ ⋅ 1 + 1 − cos(−𝑛𝜋) + cos 𝑛𝜋
2 𝑛𝜋 𝑡=−1,1 2 𝑛𝜋

𝐾 𝐾 𝑛
𝑏𝑛 = ⋅ 1 − cos 𝑛𝜋 = ⋅ 1 − −1 , 𝑛 = 1, 2, …
𝑛𝜋 𝑛𝜋
Prof. A. Yesildirek Fourier Series 19

Signals and Systems

Note
● The integral definition of the FS coefficients may be expressed using only the 1 st period signal as

1 𝑡𝑜 +𝑇𝑜 1 ∞
𝑋𝑘 = න 𝑥(𝑡)𝑒 −𝑗𝑘Ω𝑜 𝑡 𝑑𝑡 = න 𝑥1 𝑡 𝑒 −𝑗𝑘Ω𝑜 𝑡 𝑑𝑡
𝑇𝑜 𝑡𝑜 𝑇𝑜 −∞

● This looks very similar to Laplace transform

∞
𝑋1 (𝑠) = න 𝑥1 (𝑡)𝑒 −𝑠𝑡 𝑑𝑡
−∞

At 𝑠 = 𝑗𝑘Ω𝑜

So, can we use the Laplace transform of 𝑥1 (𝑡) to get the Fourier series coefficients of 𝑥(𝑡)?

Prof. A. Yesildirek Fourier Series 20

 Signals and Systems

Fourier Coefficients thru Laplace

● 𝑥(𝑡) be periodic with single period signal 𝑥1 𝑡
𝑥1 𝑡 = 𝑥 𝑡 𝑢 𝑡 − 𝑡𝑜 − 𝑢(𝑡 − (𝑡𝑜 + 𝑇𝑜 )) ∀𝑡𝑜

● Then, the FS coefficients may also be computed using

𝟏
𝑿𝒌 = L 𝒙𝟏 (𝒕) ቚ
𝑻𝟎 𝒔=𝒋𝜴𝒐 𝒌

● Provided that 𝑗𝑘Ω𝑜 is within the ROC

Prof. A. Yesildirek Fourier Series 21

Signals and Systems

Example with Laplace

● 𝑥1 𝑡 = 2 𝑢 𝑡 + 0.25 − 𝑢(𝑡 − 0.25) and 𝑇𝑜 = 1, Ω𝑜 = 2𝜋

1
● 𝑋𝑘 = L 𝑥1 (𝑡) |𝑠=𝑗Ω𝑜 𝑘
𝑇0

● 𝑋𝑘 = L{2[𝑢 𝑡 + 0.25 − 𝑢 𝑡 − 0.25 ]}|𝑠=𝑗2𝜋𝑘

𝑒 0.25𝑠 𝑒 −0.25𝑠
● 𝑋𝑘 = − |𝑠=𝑗2𝜋𝑘
𝑠 𝑠

1 𝑒 𝑗𝜋𝑘/2 𝑒 −𝑗𝜋𝑘/2 1 𝜋𝑘
● 𝑋𝑘 = − = sin ∀𝑘 ≠ 0
𝜋𝑘 𝑗2 𝑗2 𝜋𝑘 2

1 𝑡𝑜 +𝑇𝑜
● 𝑋𝑜 = ‫׬‬ 𝑥1 𝑡 𝑑𝑡 = 2 ∗ 0.5 = 1 (average)
𝑇𝑜 𝑡𝑜

Prof. A. Yesildirek Fourier Series 22

 Signals and Systems

Example

● Let 𝑥1 (𝑡) be the first period of 𝑥 𝑡 with the period 4 sec. Find the fourier series coefficients
𝑥1 𝑡 = 𝑟 𝑡 − 𝑟 𝑡 − 2 − 2𝑢 𝑡 − 3
𝜋
𝑇0 = 4 → Ω0 =
2

Prof. A. Yesildirek Fourier Series 23

Signals and Systems

𝑥1 𝑡 = 𝑟 𝑡 − 𝑟 𝑡 − 2 − 2𝑢 𝑡 − 3
Use Laplace transform
1
𝑋𝑘 = 𝐿𝑇 𝑥𝟏 𝑠 ቚ
𝑇𝑜 𝑠=𝑗Ω𝑜𝑘
1 1 − 𝑒 −2𝑠 𝑒 −3𝑠
𝑋𝑘 = −2
4 𝑠2 𝑠 𝑠=𝑗𝜋𝑘
2

3
1 1 − 𝑒 −𝑗𝜋𝑘 𝑒 −𝑗2𝜋𝑘
𝑋𝑘 = 2 −2 𝜋
4 𝜋 𝑗 𝑘
𝑗 𝑘 2
2
3
1 − 𝑒 −𝑗𝜋𝑘 𝑒 −𝑗2𝜋𝑘
𝑋𝑘 = −
−𝜋 2 𝑘 2 𝑗𝜋𝑘
1 − −1 𝑘 𝑗𝑘+1
𝑋𝑘 = +
−𝜋 2 𝑘 2 𝜋𝑘

Prof. A. Yesildirek Fourier Series 24

 Signals and Systems

𝜋
Assume that the Fourier series coefficients with Ω𝑜 = is given as
2

𝜋
−1 𝑘 − 1 𝑒 −𝑗 2 (3𝑘−1)
𝑋𝑘 = +
𝜋 2𝑘2 𝜋𝑘
𝑗𝜋 𝑗𝜋
And the kth harmonic is defined by 𝑥𝑘 𝑡 = 𝑋−𝑘 𝑒 − 2 𝑘𝑡 + 𝑋𝑘 𝑒 2 𝑘𝑡 .
Find the 5th harmonic of 𝑥(𝑡).

𝜋
−1 −5 − 1 𝑒 −𝑗 2 (−15−1) 2 1 5𝜋 + 2
𝑋−5 = + =− − =−
𝜋 2 25 −5𝜋 25𝜋 2 5𝜋 25𝜋 2
𝜋
−1 5 − 1 𝑒 −𝑗 2 (15−1) 2 1 5𝜋 + 2
𝑋5 = 2 + =− 2 − =−
𝜋 25 5𝜋 25𝜋 5𝜋 25𝜋 2

5𝜋 + 2 −𝑗𝜋𝑘𝑡 𝑗𝜋 4 + 10𝜋 𝜋
𝑥5 𝑡 = − 2 𝑒 2 + 𝑒 2 𝑘𝑡 = − 2 cos 5 𝑡
25𝜋 25𝜋 2

Prof. A. Yesildirek Fourier Series 25

Signals and Systems

Power of the Periodic Signal

● The power of the signal may be computed either in

− Time domain of average of 𝑥(𝑡) within a period

− Frequency domain using infinite sum of Fourier series coefficients 𝑋𝑘 ’s

Prof. A. Yesildirek Fourier Series 26

 Signals and Systems

Parseval’s power relation

● Let 𝑃𝒙 be the power of periodic signal with the period 𝑻𝒐

∞
1 𝑡𝑜 +𝑇𝑜 2 𝑑𝑡 2
𝑃𝑥 = න 𝑥 𝑡 = ෍ 𝑋𝑘
𝑇𝑜 𝑡𝑜
𝑘=−∞

● So, no need for integration in power calculation when the FS coefficients are known

Prof. A. Yesildirek Fourier Series 27

Signals and Systems

Let’s consider some of the properties of FS

Prof. A. Yesildirek Fourier Series 29

 Signals and Systems

Reflection and even/odd periodic signals

● Let 𝑥(𝑡) be periodic with 𝑇𝑜 and Fourier coefficients 𝑋𝑘
● Reflection
− Fourier coefficients of 𝑥(−𝑡) are 𝑋−𝑘
● Even 𝑥(𝑡)
− Fourier coefficients of 𝑥(𝑡) are real
− 𝑥 𝑡 = 𝑋𝑜 + 2 σ∞
𝑘=1 𝑋𝑘 cos(𝑘Ω𝑜 𝑡) no imaginary part

● Odd 𝑥(𝑡)
− Fourier coefficients of 𝑥(𝑡) are imaginary
− 𝑥 𝑡 = 𝑗2 σ∞
𝑘=1 𝑋𝑘 sin(𝑘Ω𝑜 𝑡) no real part

● Neither even nor odd 𝑥 𝑡 = 𝑥𝑒 𝑡 + 𝑥𝑜 (𝑡)

− 𝑋𝑒𝑘 = 0.5 𝑋𝑘 + 𝑋−𝑘 𝑋𝑜𝑘 = 0.5 𝑋𝑘 − 𝑋−𝑘

− 𝑋𝑘 = 𝑋𝑒𝑘 + 𝑋𝑜𝑘

Prof. A. Yesildirek Fourier Series 30

Signals and Systems

Example
𝑦1 (−𝑡)

Prof. A. Yesildirek Fourier Series 31

Fourier Series Properties Signals and Systems

Prof. A. Yesildirek Fourier Series 32

Signals and Systems

Shift in time property

● Let 𝑥(𝑡) be periodic with 𝑇𝑜 and Fourier coefficients 𝑋𝑘 , then time shift by 𝑡𝑜 signal

𝑦 t = 𝑥(𝑡 − 𝑡𝑜 )

has Fourier coefficients

𝒀𝒌 = 𝑿𝒌 𝒆−𝒋𝒌𝛀𝒐 𝒕𝒐

Prof. A. Yesildirek Fourier Series 33

 Signals and Systems

Examples

1 𝑒 𝑗𝜋𝑘/2 𝑒 −𝑗𝜋𝑘/2 1 𝜋𝑘
● 𝑋𝑘 = − = sin ∀𝑘 ≠ 0 & 𝑋𝑜 = 1
𝜋𝑘 𝑗2 𝑗2 2𝜋𝑘/2 2

● Let 𝑦 𝑡 = 𝐴𝑥 𝑡 + 𝐵 then

− 𝑌0 = 𝐴𝑋𝑜 + 𝐵

− 𝑌𝑘 = 𝐴𝑋𝑘 ∀𝑘 ≠ 0
● You may derive many signals from linear combinations of 𝑥(𝑡)

● 𝑦1 𝑡 = −2𝑥 𝑡 + 3 ⇒

− 𝑌10 = −2𝑋𝑜 + 3

− 𝑌1𝑘 = −2𝑋𝑘 ∀𝑘 ≠ 0

● 𝑦2 𝑡 = 4𝑥 𝑡 + 1 ⇒

− 𝑌10 = 4𝑋𝑜 + 1

− 𝑌1𝑘 = 4𝑋𝑘 ∀𝑘 ≠ 0

Prof. A. Yesildirek Fourier Series 34

Signals and Systems

With time shift property

● 𝑦3 𝑡 = −2𝑥 𝑡 + 1 + 3𝑥(𝑡 − 2) ⇒

− 𝑌3𝑘 = (−2𝑒 𝑗Ω𝑜 𝑘 + 3𝑒 −𝑗2Ω𝑜 𝑘 )𝑋𝑘

● Etc.

Prof. A. Yesildirek Fourier Series 35

 Signals and Systems

Prof. A. Yesildirek Fourier Series 36

Signals and Systems

Example

2𝐾 −𝑗𝑠𝑖𝑔𝑛(𝑛)𝜋/2
𝑋𝑛 = ቐ 𝑛𝜋 𝑒 , 𝑛 = 𝑜𝑑𝑑
0, 𝑛 = 𝑒𝑣𝑒𝑛

𝑥𝑁 𝑡 = ෍ 𝑋𝑛 𝑒 𝑗Ω𝑜 𝑛𝑡
𝑛=−𝑁

Prof. A. Yesildirek Fourier Series 37