티스토리 뷰

Key Ideas

  • discretization of time
    • samples replace idealized models
    • simple math replaces calculus
  • discretization of values
    • general-purpose storage
    • general-purpose processing (CPU)
    • noise can be controlled

 

Lesson 1.1.1: What is signal processing?

Signal?
Description of the evolution of a physical phenomenon

물리적 현상의 발생과 변화에 대한 설명 또는 묘사

Processing?
(1) Analysis: understanding the information carried by the signal
처리-분석 : 신호가 전달하는 정보를 이해하는 것
(2) Synthesis: creating a signal to contain the given information
처리-합성 : 주어진 정보를 포함하는 신호를 만드는 것

 

The Digital Paradigm

아날로그 현상을 표현하는 함수 자체를 찾는 대신, 그 함수가 만들어내는 값(values)을 기록하자

연속된 숫자로 기록된 값들로 아날로그 현상을 표현하자 = Digital

 

Key ingredients:

 

  • discrete time
  • discrete amplitude

 

What is time?

시간은 인간이 만들어낸 상상의 산물?

[...] space and time are mere thought entities and creatures of the imagination [...] They precede the existence of objects of the senses a priori. -Immanuel Kant, 1787

 

Two competing models for reality

  • 분절되지 않고 연속적으로 존재하는 실제 아날로그 현상
  • 유한수의 측정단위(finite measurement)로 분절하여 기록하여 현상을 표현 = Discrete time

 

Discrete time is very practical! Computing the average...

  • 분절되지 않은 연속 그래프 상태의 물리 현상을 평균 내려면?
    • 물리 현상을 표현하는 함수를 알아야 한다 & 적분식을 적용해야 한다
  • 유한수의 측정단위로 분절된 물리 현상을 평균 내려면?
    • 단순히 모든 측정값들을 합한 뒤 측정된 샘플 갯수로 나누어준다

 

The discrete-time model

x[n] = ..., 1.2390, -0.7372, 0.8937, 0.1798, -1.1501, -0.2642, ...

 

"n"

  • quote on time
  • n does not have a physical dimension (=adimensional)

 

Bringing the models: the founding fathers

Harry Nyquist and Claude Shannon

 

The continuous time representation and the discrete time representation are equivalent. -Harry Nyquist / Claude Shannon

 

Under appropriate "slowness" conditions for x(t) we have:

연속시간과 분절시간 간의 관계를 표현하는 수식

 

$$ x(t)= \sum^{\infty}_{n=-\infty}x[n]sinc(\frac{t-nT_s}{T_s}) $$

  • \( x(t) \) =  연속시간으로 표현되는 물리 현상
  • \( x[n] \) = 분절시간에 따라 유한개의 숫자로 측정되는 물리 현상
  • \( sinc(\frac{t-nT_s}{T_s}) \) = a linear combination of copies of a typical functionn or building block called 'sinc' shifted and scaled by the values of the discrete time sequence

 

Continuous to discrete time (and back)

Continuous \( x(t) \) -> Discrete \( x[n] \)

 

 

  1. You start with a continuous time signal
  2. You take measurements and throw away all the rest
  3. You are left with the discrete time sequence

 

Discrete \( x[n] \) -> Continuous \( x(t) \)

 

 

  1. You take copies of the \( sinc \) function
  2. You place them at each sample location scaled by the amplitude of the sample
  3. You sum all these copies of the \( sinc \) together

 

 

When can we do all this? Joseph Fourier will tell us.

  • Sampling theory에 따라 연속-분절 신호 간 변환이 가능한 조건이 정의됨
  • 연속-분절 신호 변환을 위해 Fourier analysis라는 기법이 필요함
  • 푸리에 변환(Fourier Transform)을 통해 우리는 신호가 얼마나 빠르게 움직이는지를 수치화할 수 있다
    • 신호의 속도를 알 수 있으면, 우리는 샘플링 간격(sampling interval)을 결정할 수 있다
  • Samplinng interval
    • 함수를 시퀀스로 변환할 때, samplinng theory 가정을 충족하게 하는 측정단위 사이의 간격을 의미함

 

Digital Signals - amplitude?

  • discrete time - 측정 단위, 길이, 간격에 대한 개념
  • discrete amplitude - 레벨, 상한선과 하한선, 신호값에 대한 개념

측정된 신호량을 정수(integer) 값으로 매핑하는 것은 중요하다, 왜냐하면:

 

  • storage - 정수를 저장할 수 있는 다양한 저장장치에 정보를 담을 수 있음
  • processing - 일반적인 컴퓨터 프로세서는 정수 데이터 친화적임
  • transmission - 디지털화를 통해 노이즈를 제거하고 통신채널의 capacity를 극대화하는 것이 가능해짐

 

What happens to analog signals

 

 

 

 

 

 

 

 

 

 


[코세라 디지털신호처리 Specialization 강의 내용 정리]

 

Digital Signal Processing 1: Basic Concepts and Algorithms

Module 1.1: Digital Signal Processing: the Basics

 

Learning Objectives

 

  • Classifiy discrete-time signals
  • Describe operations on signals
  • Design simple signal processing algorithms as in the example

 

https://www.coursera.org/specializations/digital-signal-processing

 

디지털 신호 처리

로잔연방공과대학교에서 제공합니다. Digital Signal Processing from theory to practice. Discover discrete-time signal and analyze them with the Fourier transform. ... 무료로 등록하십시오.

www.coursera.org

 

댓글
공지사항
최근에 올라온 글
최근에 달린 댓글
Total
Today
Yesterday
링크
«   2024/10   »
1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31
글 보관함