Forum

  • By Howard - edited: June 21, 2017

    As I mentioned in my other post, I am hoping to get information from our compliance section on this.

    In the meantime, you may find some of the information from this document useful:

    http://content.rcrwireless.com/20170517_Wi-Fi_Unlicensed_Spectrum_Report

  • Ekahau is presenting a Webinar on August 23.regarding LTE-U.

    It will include presenters from both Ruckus and Aerohive.

    The title is "Wi-Fi vs LTE-U: Here’s to the Mess We Made"

    The  sign-up link is at:   http://wifi.ekahau.com/wi-fi-vs-lte-u

    Really looking forward to it.

  • Regarding RED.

    I was directed to a couple of documents on the ETSI site, and found two documents, one each for 2.4 GHz and 5 GHz.  Unfortunately, I don't have a lot of spare time right now, so I really can't do a thorough analysis of them - I'll have to let let Compliance worry about it.

    It appears that in 5 GHz  that the main concern is with DFS channels - as you might expect.

    For both 2.4 and 5 GHz there looks to be an emphasis on "blocking channels", which seems to be a very different viewpoint from the "Energy Detect" levels I am used to thinking about.    

    This looks much more like Adjacent Channel Rejection testing, but instead of using "compliant" signals (DSSS or OFDM), they only use CW.

    I don't think I've seen any Regulatory Body rules covering ACR or ED tests for Wi-Fi before - only IEEE.

    My Anritsu WLAN Test Set can perform this type of test, but Anritsu specifies modulated signals from the interferer - which I could do since the interferer is an external sig-gen.

    I ran only a few ADJ-CH rejection tests many years ago, so I don't have enough experience to say how the different requirements might affect a radios performance. 

    It does look like these tests, what ever you want to call them, might mitigate some of the interference caused by LTE-U.    I guess time will tell.

    For 2.4 GHz, see:

     http://www.etsi.org/deliver/etsi_en/300300_300399/300328/02.01.01_60/en_300328v020101p.pdf

    For 5 GHz band, see:

    http://www.etsi.org/deliver/etsi_en/301800_301899/301893/02.01.01_60/en_301893v020101p.pdf

    Cheers

  • Thanks for the reading material Howard.  This is good to have.  The comment in section B.7.3 that the test signal is continuously modulated "LTE-type" signal does give me pause about the realism of the test conditions too.    

    To be continued...

    Cheers,

      David

  • By Howard - edited: June 23, 2017

    You're right.  

    In section B.7.2 it also says:

    "The OFDM test signal shall consist of a continuous sequence (100 % duty cycle) of OFDM data symbols as defined in IEEE 802.11™-2016 [9], clause 17. This implies that the OFDM test signal does not contain any OFDM PHY preambles as defined in IEEE 802.11™-2016 ..."  (my emphasis)

    How many people have the ability to do that?.   I used to use another WLAN test set.  

    Unless you have an AWGN, you'll be out of luck - another thing to add to the budget.

    Also, the FCC never really cared about OFDM Spectral Masks per se, except at the extremes of the bands.   These are now more carefully defined.    I know for a fact that many radios in the last 8 years, or so, did not meet the mask limits (and there are probably others I don't know about).   Unlike stairs, it's the 2nd step down that is the killer.

    I'm almost surprised they didn't include EVM limits !   :-)

  • https://www.engadget.com/2017/06/26/t-mobile-launches-unlicensed-lte/. And so it begins.  LTE license assisted in this reference.

  • After further evaluation of the ETSI document for 5 GHZ devices, it appears that the issues involved  with using an AWG is much more complicated than what I first thought (in section B7.2.)    This non-802.11 OFDM compliant signal must also be interpreted, or at least recognized, by the WLAN or LTE (DUT) device.    Which means custom firmware is required in the DUT  and I was hoping to get away from needing custom firmware while testing.

    Luckily, there does not seem to be a similar requirement for testing in the 2.4 GHz band - i.e no special DSSS signal is required.   Of course, there are no LTE signals in the 2.4 GHz band, but I bring it up for completeness.

    In both bands, the interfering signal is a CW signal - not one which conforms to the modulation format of either LTE or WLAN.

    This last fact is interesting, and is something I can manage in my lab.    I should be able to perform the 2.4 GHz test if I should decide to do so.  

    I will not be able to perform the ETSI 5 GHz Receiver Blocking tests, but perhaps I can find a correlation point where I can approximate them using normal OFDM signals.

    Darn.   I guess our Compliance group will be the only ones that can run the 5 GHz tests.

  • How disappointing - ETSI is as slow as the IEEE.

    The latest version of the 2.4 GHz ETSI standard is dated 1/13/2017.   However, according to this ETSI status page, and links:

    https://portal.etsi.org/webapp/workprogram/Report_WorkItem.asp?WKI_ID=51206

    It is not due to be published until 8/21/2018.

    So it looks like the 11/2016 version is all we have to go by for now.

  • It does seem like we are getting a bit into a draw your curves then plot your points type of situation for the tests.

    How about plan C for a standards body that would make a difference?  Are there any in Asia that may help address this problem?

    If any of this is pay for view I might me able to get my work to pay for it, with proper business justification of course. 

  • By Howard - edited: August 17, 2017

    Next Wednesday, Aug. 23  is the date for the Ekahau  "Wi-Fi -vs- LTE-U" webinar.

    The first line of their email announcement reads:

    Fact: Cellular is going to hurt your Wi-Fi.

    You can sign up for the webinar at:

    http://wifi.ekahau.com/wi-fi-vs-lte-u

    See the Webinar outline on the sign-up page.

    Enjoy

Page 2 of 4