Talk:Constellation diagram

Latest comment: 17 years ago by Alinja in topic Changes 3rd Jan 2007

Changes 3rd Jan 2007

edit

Reasons for reversion:

  • Analogue signals may also be baseband, therefore to describe constellation diagrams as "a representation of a baseband signal" is less correct than "a digital modulation scheme".
  • QAM is not the only way to generate a modulation scheme representable by a constellation diagram. At its most general, a constellation diagram represents the phase and amplitude of symbols, which could be generated in other ways.
  • A constellation diagram is not a histogram.
  • "This principle of using orthogonality is the base of quadrature modulation." This is confusing, at best. No prior mention of orthogonality was given; plus it implies that orthogonality in general is sufficient for quadrature modulation.
  • Description of how complex numbers are used to modulate cos and sine waves was removed.
  • "As the symbols are represented as complex numbers, they can be visualized as points on the complex plane." is a tautology.
  • "Plotting several symbols this way in one diagram produces the constellation diagram. Also a diagram of the ideal positions in a modulation scheme can be called a constellation diagram." is confusing, IMO. This could be re-phrased along the lines of "Constellation diagrams are often used to depict ideal symbol locations, and the transitions between them. A constellation diagram may also be used to depict actual symbol locations, which may differ due to noise..." But I'm too tired right now to do it properly.

Oli Filth 23:20, 3 January 2007 (UTC)Reply

True, I ended up writing some things less clearly than I intended. But still:
  • My description didn't include a mention about digital modulation, which a constellation diagram obviously requires. But constellation certainly is used to describe signals in addition to just the scheme.
  • The definition should be reworded anyway, because I didn't think of true phase modulation.
  • A lot of measurement equipment displays the constellation as a coloured 2-dimensional histogram. But true, it would be more intuitive to just describe it as a basic scatter diagram here.
  • Orthogonality (in the vector space sense) of carriers is sufficient to transmit a 2-dimensional vector and therefore for transmitting a complex number (and sufficient for constellation representation). But it is probably a too advanced concept here and not everyone who reads this is aware of the term. A more basic explanation is better here.
  • The description of the modulation is in the 2nd paragraph. While it is necessary background for this article, the details of it should be in QAM/PSK articles anyway.
  • Tautology maybe, but an essential part of the article is in that sentece. It should be clearly worded, but can you suggest a better one?
  • Constellation diagram doesn't include the transitions (xy- or IQ-diagram includes them) and there's no reason to repeat the reasons for noise etc. Maybe: "
In the original article "By choosing ... in this way" and "are a convenient representation of the two axes" are vague, non-cohesive and maybe don't even make sense. "The amplitude of the cos wave (or sin wave) is set to the absolute value of the imaginary part of the symbol to be transmitted and the amplitude of the sin wave (or cos wave) to the absolute value of the real part" mixes sin and cos, and the part about absotule values is incorrect. For these reasons I think this article needs a rewrite. Alinja 11:48, 4 January 2007 (UTC)Reply