information is a subtle mixture of predictability and surprise

Information Theory - Mathematical Theory of Communication
  • studies the quantification of information both in storage and communication
  • originally proposed by Claude Shannon in 1948 to find fundamental limits on signal processing and communication operations such as data compression, in a landmark paper titled A Mathematical Theory of Communication

Communication System

A communication system has 5 components:

  • information source - produces a message or sequence of messages to be communicated to the receiving terminal
  • transmitter - operates on the message in some way to produce a signal suitable for transmission over the channel
  • channel - the medium used to transmit the signal from transmitter to receiver
  • receiver - operates on the signal to produce the message
  • destination - is the entity for whom the message is intended for

Communication System - Categories/Classes

Message Type
  • discrete message - the message is a sequence of discrete symbols (e.g. sequence of letters)
  • continuous message - the message is treated as a continuous function (e.g. voice)
Signal Type
  • discrete signal - the signal is a sequence of discrete symbols (e.g. bits in an IP network)
  • continuous signal - the signal are treated as continuous functions (e.g. electromagnetic waves, specifically radio waves)
Communication System Type (Message & Signal)
  • discrete system - both the message and signal are discrete (e.g. telegraph; messages are sequences of letters, signals are sequences of dots, dashes, and spaces of dots, dashes, and spaces)
  • continuous system - both the message and signal are discrete (e.g. radio or television)
  • mixed system - message and signal is mixed (e.g. PCM transmission of speech)
Channel Type
  • noiseless channel - an idealistic channel in which no information is lost, corrupted, nor duplicated
  • noisy channel - a realistic channel in which information can be lost, corrupted, or duplicated

Message Source Statistical

𝑆 is the size of possible symbols such as (letters, words, etc)

  • zero-order - each message symbol has an equal probability of being produced
  • first-order - 𝑆⁰=1 state - unigrams - each message symbol has
  • second-order - 𝑆¹ states - bigrams -
  • third-order - 𝑆² states - trigrams -
  • nth-order - 𝑆𝑛 states - n-grams -

Subpages

Resources