96 lines
7.2 KiB
Typst
96 lines
7.2 KiB
Typst
#import "@preview/wordometer:0.1.4": word-count, total-words
|
||
|
||
#set page(
|
||
paper: "a4",
|
||
numbering: "1",
|
||
margin: (top: 2.5cm, left: 2.5cm, right: 2.5cm, bottom: 2cm)
|
||
)
|
||
|
||
#set text(
|
||
font: "Times New Roman",
|
||
size: 12pt,
|
||
)
|
||
|
||
Marius Drechsler\
|
||
Process Essay\
|
||
May 17th, 2025
|
||
|
||
#align(center, text(size: 17pt, weight: "bold")[
|
||
*The Digital Journey of Your Voice*
|
||
])
|
||
|
||
#set align(left)
|
||
#set par(
|
||
justify: true,
|
||
leading: 2em,
|
||
spacing: 2em,
|
||
first-line-indent: (amount: 3em, all: true)
|
||
)
|
||
|
||
#show: word-count
|
||
|
||
Have you ever wondered what happens with your voice when you are talking to someone on the phone?
|
||
From the instant the soundwaves leave your throat until they reach the ear of the person you are talking to,
|
||
a series of analog and digital processes collaborate to carry your message.
|
||
In fact, this whole process can be broken down into three major steps -- sampling, quantisation and modulation.
|
||
In the course of this essay, we will investigate each of these steps in more depth to understand how modern
|
||
communication works on a technical level.
|
||
|
||
To start, we will take a closer look at the analoue signal that reaches your phone's microphone.
|
||
Every sound wave, like your voice or the tone of a guitar string, is so called time and value continuous.
|
||
That means, such a signal has an infinitely accurate value at each imaginable point in time.
|
||
However, an electronic device, for example a computer or a phone, cannot understand such an analogue signal, thus we have to first transform it into some kind of electrical signal the device can understand.
|
||
In general, we can assume that an electrical device can only process time and value disteet signals.
|
||
To transform our original continuous signal into its discreet or rather digital representation, we can make use of the sampling and
|
||
quantization steps in our communication process.
|
||
|
||
In the sampling process, the analogue signal is transformet into a time discreet and value continuous signal.
|
||
Conceptually, an analog-to-digital converter (ADC) takes rapid "snapshots" of the amplitude of the input signal at uniform intervals and records each reading.
|
||
The rate at which these snapshots occur is called sampling frequency.
|
||
Ideally, we do want to maximize the timespan between each of these snapshots, using the lowest possible sampling frequency so to speak.
|
||
The Nyquist Frequency defines this lowest possible sampling frequency as double the frequency of the originating signal @shannon.
|
||
For example, if the frequency of our original signal is 1 MHz, the analog-to-digital converter will need to take a snapshot of the signal at a frequency of at least 2 MHz.
|
||
At this moment, our signal now is time-discreet and value-continuous.
|
||
To convert these time-discreet values into real bits and bytes we will make use of the quantizer in the next step.
|
||
|
||
Quanization describes the operation of transforming a continuous value into a discreet form.
|
||
Imagine placing random dots in a row on a piece of paper.
|
||
A simple quantization process now would be to take a ruler and for each point record which is the next highest marking on the ruler.
|
||
In the same sense, the quantizer of an electronic device maps the continuous input vaulues to predefined codewords -- a collection of bits (for example "000", "110" or "011").
|
||
Our signal has now fully arrived in the digital domain.
|
||
// Note: Maybe expand more what happened up until now
|
||
Up to this point we have completeley encoded our analogue message digitally.
|
||
During the next steps, the digital signal will be further processed and prepared for transmission.
|
||
|
||
A digital signal in its raw form is very inefficient to transmit because of the limited underlying bandwidth.
|
||
Because we need to transport our message over some kind of communication channel, the amount of information we can transport in a fixed period of time is limited.
|
||
The first step in solving this issue is using compression by removing irrelevant and redundant information from the signal.
|
||
A popular example for a compression method is called Huffman Coding @huffman.
|
||
Conceptually, the Huffman Code consists of multiple codewords of varying length where symbols with a higher probability of occurrence are assigned to the shorter codewords, thus reducing the overall size of the message.
|
||
Since the assignment of symbols to codewords is based on their probability of occurrence, this method of compression requires information about the statistics of the incoming symbols.
|
||
If these statistical information are not known, other compression methods such as the Lempel–Ziv–Welch algorithm can be used.
|
||
Furtheremore, compression algorithms specifically tailored for different signal sources can be used, for example PNG for pictures, MP3 for audio or MPEG for video signals.
|
||
With the analogue message digizited and compressed for easier transport over our communication channel, we will now need to prepare our message on a physical level for it to be able to be transmitted using a radio wave or a data cable.
|
||
|
||
Currently, the message to be transmitted can be represented as a set of codewords like "00 01 10 11".
|
||
To prepare our message digital and compressed message for transmission over a physical channel, digital modulation -- like amplitude modulation -- is used.
|
||
This works by defining a specific signal amplitude for every possible codeword, which is called Amplitude-Shift Keying (ASK) @modulation.
|
||
The simplest form of ASK is called On-Off Keying (OOK), where we will either transmit a wave -- and signaling a binary 1 -- or not transmit anything -- and signal a 0.
|
||
In our example we may define four sine functions with varying amplitutes as the modulated signal that is being transported over the physical communication channel.
|
||
Because we defined four different Amplitude-Shifts, this type of modulation is called "4 ASK" @modulation.
|
||
Signal modulation is not limited to changing the ampliude of our transmission signal, thus we can also alter the phase or frequency of the signal.
|
||
Depending on the type of communication channel we may want to choose a different modulation type using either one or a combination of different modulation parameters.
|
||
For example, a popular modulation type that uses a combination of amplitude shifts and phase shifts is called "Amplitude-Phase-Shift Keying (APSK) @modulation.
|
||
Using modulation, we prepared our signal on a physical level to instruct a communication interface -- like an antenna or an optical transmitter -- to finally transmit our message.
|
||
|
||
As final step, the receiver of the message has to process the received signals in exactly the reverse order to create a comprehensible message.
|
||
The most important prerequisite for this to work is that both sender and receiver have agreed on the same transmission and reception conditions.
|
||
The receiver will first need to use the correct modulation type to convert their received signal back to a set of codewords.
|
||
Going on, they will decompress the message based on the used compression algorithm and use a Digital-to-Analog Converter (DAC) to transform the digital message back into sound waves which will be output by the speaker of their phone.
|
||
This whole process now happens at such a high speed that makes it possible for us to talk to a person on the other side of the world.
|
||
|
||
//Essay has a total of #total-words words.
|
||
|
||
#pagebreak()
|
||
|
||
#bibliography("./bibliography.bib", style: "ieee", title: "References")
|