Talk:Stokes parameters

Merge?

It seems to me that Stokes vector and Stokes parameters should be merged into a single page. Does this seem reasonable, and if so which of the two names should be chosen for the combined page?--Srleffler 21:54, 15 December 2005 (UTC)

The two are very closely related. When using the historical I, Q, U, and V, they are usually referred to as parameters. When using S in a matrix calculation, it is a vector. When talking about the information content of say S3 it can be called a parameter again. I think one term should redirect to the other term. I am comfortable with it going either way. Rodney Meyer Rodney 16:43, 16 December 2005 (UTC)

I agree, maybe the chosen name should be something like: "Stokes polarization description". It should be noted that Stokes parameters are parts of Stokes Vector and called just Stokes Vector elements.
Emil Kaczmarek

No, it should definitely be named either Stokes vector or Stokes parameters. Wikipedia articles are always named by the most common (linkable) name available, if possible.--Srleffler 11:40, 24 April 2006 (UTC)

Currently, the nomenclature is turning to Stokes parameters only. This is because Stokes parameters do not form a proper vector space, i.e., you can do a linear transformation on a Stokes vector which results in a non-physical Stokes vector...hence the name Stokes parameters. — Preceding unsigned comment added by 68.104.252.75 (talk) 23:27, 27 January 2015 (UTC)

There's no page for "Jones vector" because it falls under "Jones calculus". The same should be true for Mueller calculus: We don't need extra pages for Stokes vector and Stokes parameter.207.91.187.66 (talk) 16:54, 2 May 2024 (UTC)

Corrections

I removed erroneous equations about eccentricity/ellipticity. Not sure how they got there in the first place. Gnixon 13:55, 22 August 2006 (UTC)

I think there is a mistake on the formula to calculate V Stokes parameter form jones parameter. a left circular polarization being V>0 and described in jones in something of the form (1;i), the conversion should be I think 2Im(Ex*.Ey) and not 2Im(Ex.Ey*). Can someone confirm that? Nicolas Lefaudeux, 22 september 2008

wrong sphere illustration

The figure illustrating the poincare sphere is not correct in the sence that the sphere is not correctly displayed/calculated in this frontal projection! A sphere is somehow stretched in such a projection. —Preceding unsigned comment added by 153.96.200.9 (talk) 15:58, 14 July 2010 (UTC)

Probable Error in Equation for V Stokes parameter in fixed x,y basis

I believe there is an error in the equation for the Stokes parameters in the fixed x,y basis. There should be a minus sign in the expression for the V

V = -2 Im(E_x,E_y*)

See equation 2.51 page 45 Polarimetric Radar Imaging, from Basics to Application, J.S. Lee and Eric Pottier, 2009, CRC Press. I confirmed the calculation in this reference.

Actually this depends on the phase convention used for Jones matrices. If, like in Born and Wolf, you use the decreasing phase convention, then you will get

V = 2 Im(E_x,E_y*)

If you use the increasing phase convention (Collett and others) then you get your answer. Since the Jones calculus article uses the decreasing phase convention, this should be changed back to the Born and Wolf Definition. — Preceding unsigned comment added by 68.104.252.75 (talk) 23:36, 27 January 2015 (UTC)

Yes, I also find some sign problems in those formula in the "Representations in fixed bases" section.

According to my derivation, I get

For (x,y) basis:

${\displaystyle$ :{\begin{aligned}:I&\equiv |E_{x}|^{2}+|E_{y}|^{2}\\:Q&\equiv |E_{x}|^{2}-|E_{y}|^{2}\\:U&\equiv |E_{a}|^{2}-|E_{b}|^{2}=E_{x}E_{y}^{*}+E_{x}^{*}E_{y}=2\Re {(E_{x}E_{y}^{*})}\\:V&\equiv |E_{r}|^{2}-|E_{l}|^{2}=-i(E_{x}E_{y}^{*}-E_{x}^{*}E_{y})=2\Im {(E_{x}E_{y}^{*})}:\end{aligned}}:}

For (a,b) basis:

${\displaystyle$ :{\begin{aligned}:I&\equiv |E_{x}|^{2}+|E_{y}|^{2}=|E_{a}|^{2}+|E_{b}|^{2}\\:Q&\equiv |E_{x}|^{2}-|E_{y}|^{2}=E_{a}E_{b}^{*}+E_{a}^{*}E_{b}=2\Re {(E_{a}E_{b}^{*})}\\:U&\equiv |E_{a}|^{2}-|E_{b}|^{2}\\:V&\equiv |E_{r}|^{2}-|E_{l}|^{2}=i(E_{a}E_{b}^{*}-E_{a}^{*}E_{b})=-2\Im {(E_{a}E_{b}^{*})}:\end{aligned}}:}

For (l,r) basis:

${\displaystyle$ :{\begin{aligned}:I&\equiv |E_{x}|^{2}+|E_{y}|^{2}=|E_{l}|^{2}+|E_{r}|^{2}\\:Q&\equiv |E_{x}|^{2}-|E_{y}|^{2}=E_{l}E_{r}^{*}+E_{l}^{*}E_{r}=2\Re {(E_{l}E_{r}^{*})}\\:U&\equiv |E_{a}|^{2}-|E_{b}|^{2}=i(E_{l}E_{r}^{*}-E_{l}^{*}E_{r})=-2\Im {(E_{l}E_{r}^{*})}\\:V&\equiv |E_{r}|^{2}-|E_{l}|^{2}:\end{aligned}}:}

Xiangdong Zhang (talk) 14:54, 6 March 2026 (UTC)

Make sure you are using the correct sign convention. As with many areas of physics, there are arbitrary choices of sign that must be made to derive the formulas. If you use a different sign convention, you get different signs on various terms in the formulas. The article needs to remain consistent. Don't try to fix formulas unless you understand the sign convention used and ensure that you are consistent with the rest of the article. If you think formulas in the article are inconsistent with one another or with the stated sign convention, explain that.

Original research is not acceptable on Wikipedia. Articles need to be written with reference to reliable sources, not an editor's own work. I've seen too many errors introduced in articles by editors attempting to "fix" formulas based on their own derivations. If we need to validate formulas in the article, we should identify a source that uses the same sign convention as the article and check the formulas against that, not against one editor's own derivation.--Srleffler (talk) 00:49, 8 March 2026 (UTC)

Yes, that's why I didn't edit the article directly, but post my result in this talk. I am not familar with the reference of Stokes parameter, I just find some inconsistency in this wiki, based on my own derivation. I cannot upload PDF file so far since this is a new account, please wait some days and I will upload my full derivation for the reference of other editors. Xiangdong Zhang (talk) 01:20, 9 March 2026 (UTC)

Please see this PDF file for detailed derivation: [Verification of Wikipedia Page Stokes parameters.pdf]

File:Verification of Wikipedia Page Stokes parameters.pdf

Verification of Wikipedia Page Stokes parameters

Xiangdong Zhang (talk) 08:36, 12 March 2026 (UTC)

Pick one notation, please

Do we really need 2 different names for each Stokes parameter? (S1,S2,S3,S4) and (I,Q,U,V). It's redundant, but also very confusing because it looks like the first half of the article was written by someone who prefers (S1,S2,S3,S4) notation whereas the latter half was written by someone who prefers the other notation.162.246.139.210 (talk) 21:25, 19 May 2023 (UTC)

We certainly have to explain both notations, since readers may have seen one or the other elsewhere. Agree that it would be better if the article were consistent in its own notation, but I'm not going to put in the time to go through and fix it. If you have time, feel free to do so.--Srleffler (talk) 16:05, 20 May 2023 (UTC)

{Sn} vs {I,Q,U,V} is a bit confusing, but at least it is explained well and easy to translate. However, the whole article is riddled with different notations that are not related to each other. In fact, it is two or three articles conjoined, not merged, each with their own independent explanations and notations. For example, A, B vs Ea, Eb or Ex, Ey, Chi, Psi vs theta, etc. I feel that the article should be rewritten, combining the notations, and pulling in the best arguments from each take. Then there could be a small section explaining the alternate notations. Chris2crawford (talk) 11:57, 1 August 2023 (UTC)

That sounds like a good plan.--Srleffler (talk) 15:55, 6 August 2023 (UTC)

The definition of V in Figure and text mismatch

In the "Definitions" section, we defined $V\equiv |E_{r}|^{2}-|E_{l}|^{2}$, which means a left-hand circular polarized light should have $V=-1$. But in the figure before "Definitions" section and the figure before "Representations in fixed bases" section, they show that the sign of V for left-hand circular polarized light is "+V". This seems to be a mismatch. I suggest to follow the definition $V\equiv |E_{r}|^{2}-|E_{l}|^{2}$ and update the figure accordingly. Xiangdong Zhang (talk) 14:00, 6 March 2026 (UTC)