A Very Basic Introduction To

Time/Frequency Domains
Particle
March 10, 2004

Abstract
A very brief introduction to waves, terminology, time/frequency
domains, with a bit of mention of various transforms.

1 Introduction
In the context of communications, a signal is basically some information
somehow encoded as a wave. Everything travels as a wave, so let’s look at
some of the simple ideas of that. Figure 1 presents a basic Sin wave.
The vertical axis is the volume, or air-pressure, or strength, or energy,
etc., depending on what this wave is. For example, if this is an ocean wave,
then the vertical axis may be the wave height. If this is a sound wave in the
air, then it is the air-pressure. If it is the sound wave in a wire, then it’s the
voltage.
The horizontal axis (here depicted in terms of π) often represents the time
or distance (or space) that the wave traverses in.
Waves generally have a period, which can be viewed as the time/distance
of when the wave repeats itself—starts the next iteration. In Figure 1 the
period is 2π.
A closely related property of the period is the frequency. Frequency is
the speed at which the wave cycles its periods, and is always expressed in
proper speed units, as in ‘cycles per second’ (commonly known as Hz), and
‘kilo-cycles per second’ (commonly known as kHz)1 .
1
Naturally ‘kilo’ stands for 1000, so if something is 8kHz, then it’s 8000Hz.

1
 1

Amplitude

Voltage/Air Pressure
0.5

−0.5

−1
Period

0 π/2 π 3π/2 2π 5π/2 3π

Time/Distance

Figure 1: Analog Sin Wave.

2 Time Domain

1
Voltage/Air Pressure

0.5

−0.5

−1
0s 0.25s 0.5s 0.75s 1s
Time

Figure 2: 4Hz Sin Wave; sin(2π4t)

Figure 2 illustrates a Sin wave at 4Hz. Notice that the horizontal axis is
now time, and is expressed in seconds. You can count the number of times
the wave cycles, and you’ll see that it’s 4, ie: 4Hz.

2
 Given the frequency, you can easily find the wave period. For example,
if the wave cycles itself 4 times per second, then the period is 14 seconds,
or simply 1/F requency. Similarly, knowing the period, you can find the
frequency.
Another closely related value is the wavelength, which is basically the
distance occupied by the wave period. Given the speed of propagation (light
in a vacuum, or sound in the air, etc.), we can calculate that too. For
example, light/electricity travel around 3 × 108 meters per second. Let’s
imagine that Figure 2 represents electricity in a wire, then the wavelength
is just 3 × 108 ÷ 4 or just 7.5 × 107 meters, or simply velocity ÷ f requency.
Obviously you can find velocity if you know the wavelength and frequency.

1
Voltage/Air Pressure

0.5

−0.5

−1
0s 0.25s 0.5s 0.75s 1s
Time

1
Figure 3: 12Hz Sin Wave; 3 sin(2π12t)

Continuing on with the discussion, Figure 3 shows a wave of 12Hz, and

with 13 amplitude. Notice how the period got smaller as the frequency in-
creased.

3 Frequency Domain
Figure 4 illustrates the idea of taking two waves and adding them together.
In this case, we’ve added waves from Figures 2 and 3.
There are primarily two ways of viewing any type of a wave; in the time
domain, or in the frequency domain. The frequency domain of Figure 4 might

3
 1

Voltage/Air Pressure
0.5

−0.5

−1
0s 0.25s 0.5s 0.75s 1s
Time

Figure 4: 4Hz + 12Hz Sin Wave.

look something like Figure 5.

0.75
Amplitude

0.5

0.25

0
0Hz 4Hz 8Hz 12Hz 16Hz 20Hz 24Hz 28Hz 32Hz
Frequencies

Figure 5: Frequency Domain of 4Hz + 12Hz Sin Waves.

4
4 Applications
It turns out that viewing waves in frequency domain is usually a lot more
useful than viewing them in the time domain.

5 Discrete Fourier Transform

N −1
hk e2πikn/N
X
Hn =
k=0

6 Inverse Discrete Fourier Transform

1 NX
−1
hk = Hn e−2πikn/N
N n=0

7 2D Discrete Cosine Transform

JPEG, etc:
N −1 N
" # " #
1 X X −1
(2x + 1)iπ (2y + 1)jπ
DCT (i, j) = √ C(i)C(j) P ixel(x, y)COS COS
2N x=0 y=0 2N 2N

Where C(x) = √1 if x is 0, else 1 if x > 0.

2

8 2D Inverse Discrete Cosine Transform

1 NX−1 N
" # " #
X −1
(2x + 1)iπ (2y + 1)jπ
P ixel(x, y) = √ C(i)C(j)DCT (i, j)COS COS
2N i=0 j=0 2N 2N

Where C(x) = √1 if x is 0, else 1 if x > 0.

2

9 Conclusion
In class: How JPEG works.