Talk:Fading

Latest comment: 5 years ago by KorgBoy in topic Coherence time definition

Coherence time definition

edit

Currently, the article page definition of 'coherence time' seems really confusing. It is currently described (on the article page) as: "The coherence time is a measure of the minimum time required for the magnitude change or phase change of the channel to become uncorrelated from its previous value." KorgBoy (talk) 05:17, 7 December 2018 (UTC)Reply



Contradicting definitions of slow and fast fading

edit

The article defines slow and fast fading in a contradictive manner. The doppler spread section need to be improved. Mange01 11:57, 17 November 2006 (UTC)Reply

This sounds like the perfect opportunity for you to be bold and help improve the article!
Atlant 14:06, 17 November 2006 (UTC)Reply
I would like to remove the doppler spread related definition, because I do not recognize it from the literature. Mange01 01:37, 27 November 2006 (UTC)Reply

Why the slow fading is mixed up with the shadow fading. These things are completely different. Large scale fading, shadow fading is not relative to coherence time of the channel. This is usual mix up in terms that people make also in literature as well —Preceding unsigned comment added by 192.100.116.143 (talk) 07:03, 18 September 2008 (UTC)Reply

Equalizers as a diversity technique

edit

I notice Mange01 has re-inserted the following addition to the list of diversity techniques in the Mitigation section of the article:

  • Equalizers, to combat phase shift, frequency selective fading and to some extent also inter-symbol interference

I have no objection to including a mention of equalization somewhere in the article, as it performs a useful receiver function in both fading and AWGN channels. However, I feel that the location of its current mention -- as a short bullet point in a list of methods to specifically improve fading performance through diversity -- is misleading.

Specifically, as currently written the "Mitigation" section is used to emphasize that fading channels fundamentally have much poorer error performance than AWGN channels due to the fact that the channel gain is random and subject to deep fading events. Performance of all receivers becomes limited by the probability of a fading event in a single channel pat h, independent of what receiver algorithms are used. This causes the probability of error to decay as 1/SNR rather than exp(SNR). (A good reference for the derivation is the chapter on diversity in Fundamentals of Wireless Communication by Tse and Viswanath.)

This is typically combated by using diversity to create multiple parallel and independently fading channels to reduce the chance of the whole channel experiencing a fade. If the probability of a deep fade on a single channel is Pf, then the probability of a deep fade affecting all N independent channels is reduced to Pf^N. The corresponding effect is to increase the error exponent from SNR^(-1) to SNR^(-N).

All the techniques currently listed in that section (OFDM, Rake, STC, MIMO) specifically address the random nature of the channel gain by using such diversity techniques. Equalization, however, does not increase the diversity of the channel, so error performance is still dominated by the probability of a fade in a single channel path. At best it improves the constant coefficient of the error probability, and not the error exponent.

Mange01, if you think a specific mention of equalization in the fading context would be beneficial to the article, then I suggest moving it from the current location in the list of diversity techniques and adding a clarification of the difference between equalization and the list of diversity techniques in that section would be very helpful. In its current state, the section is misleading because it suggests that equalizers can also provide robustness to deep fading events. Elch Yenn 14:39, 26 July 2007 (UTC)Reply

First, thankyou Elch Yenn for improving the quality of the article.
My students often have problems understanding that for a AWGN channel with flat fading, adaptive equalization is normally omitted in simulation models, while equalization may be necessary for a dispersive fading channel. If it is important for you, please move the equalization paragraph for example to the section about freq sel fading, or stress that equalization is not a diversity technique.
Still the article needs to be supplemented by a discussion about time spreading, inter-symbol interference and the effect of doppler shift combined with fading. The section about dispersive fading models could be extended, and concrete examples of "standard models" for diferent scenarios with numerical parameter values could be presented. The relation between coherence bandwith and coherence time could be mentioned.
20:50, 26 July 2007 (UTC)~
Thanks for the feedback and suggestions! I went ahead and moved the equalizer mention to the section on frequency-selective fading; in light of your comments that seems like a good fit. Also following your suggestion, I've also added a note regarding time dispersiveness and ISI caused by frequency-selective fading. Since this effect is a general property of frequency-selective channels and not specifically a consequence of fading, my first thought would be that if you think more depth would be useful on dispersive channels and ISI mitigation techniques, it might be a good idea to start an article on frequency-selective channels and link to that.
I agree that a discussion relating coherence bandwidth and delay spread, along with coherence time and Doppler spread, would probably be good too -- there was a brief mention in an earlier version of the article that was lost when I tried to revise for readability, for which I apologize. Please feel free to add in such a discussion; I'll also try to get to it soon if you don't. Elch Yenn 22:35, 26 July 2007 (UTC)Reply

Shadowing

edit

This page is a redirection for shadowing, but it is not explained what shadowing is. In stead, this page refers to itself to explain shadowing. WKW 130.89.14.44 (talk) 10:37, 14 May 2009 (UTC)Reply