tag:blogger.com,1999:blog-1988432060681510848.post765323678324281724..comments2024-03-25T23:51:47.067-05:00Comments on Revolution Wi-Fi: Optimized Roaming, RSSI Low Check, RX-SOP, Oh My!Andrew von Nagyhttp://www.blogger.com/profile/12658799453646609565noreply@blogger.comBlogger9125tag:blogger.com,1999:blog-1988432060681510848.post-84982582809246428992014-09-04T15:00:18.784-05:002014-09-04T15:00:18.784-05:00Hi Rupert,
I would think it does impact RRM since ...Hi Rupert,<br />I would think it does impact RRM since RRM neighbor packets from other APs might be ignored if they are below the RX-SOP threshold and therefore not demodulated. It would likely impact both transmit power control (TPC) and dynamic channel assignment (DCA). However, we need more information from Cisco as to how RX-SOP impacts RRM to be certain.<br /><br />Cheers,<br />AndrewAndrew von Nagyhttps://www.blogger.com/profile/12658799453646609565noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-49432794051021011542014-09-04T11:32:04.737-05:002014-09-04T11:32:04.737-05:00So, that RX-SOP impact RRM algorithms at all?So, that RX-SOP impact RRM algorithms at all?Anonymoushttps://www.blogger.com/profile/03107779763190028272noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-40597420688810459642014-09-02T16:00:42.248-05:002014-09-02T16:00:42.248-05:00RX-SOP does indeed happen at CCA, so the solution ...RX-SOP does indeed happen at CCA, so the solution is quite simple and effective. If the received frame falls below the RX-SOP threshold, it is discarded as noise and is not even demodulated. We spend no time processing the packet.<br /><br />The No Strings Attached Show recently published results of their independent testing of this feature here: http://nostringsattachedshow.com/RXSOP/<br /><br />And Cisco posted an RX-SOP deep dive video from WFD7 here: https://www.youtube.com/watch?v=p5nx4L3RIVk<br />The question of "how is RX-SOP implemented" is addressed at 15:52 in the video.Anonymoushttps://www.blogger.com/profile/08072823562640273559noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-49950917273066356052014-08-22T04:11:47.228-05:002014-08-22T04:11:47.228-05:00Great article @Mike, your simple yet informative e...Great article @Mike, your simple yet informative explanation made things clear. I am interested to know when the RX-SOP was introduced the first time. You mentioned "7.2 code", but it would be more interesting to know the exact date. Maybe @Travis can help to clear this out.<br />Thank you.Anonymoushttps://www.blogger.com/profile/04515024765075255791noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-77645921574587727662014-08-18T11:38:33.960-05:002014-08-18T11:38:33.960-05:00OK, so I need to back peddle a bit and correct mys...OK, so I need to back peddle a bit and correct myself. I did some digging on internal documentation on this, to verify my understanding. RXSOP seems modify the CS value for the AP, which will allow it to transmit down stream, provided it can pass the ED portion of CCA. I've had a hard time finding explicit docs saying this, but based on other ones I can find, this is what I have to assume it's doing. My misunderstanding was that SOP was completely separate from the CCA ED/CS portion and really only addressed the incoming frames and allowed the AP to disregard the ones that didn't meet the defined threshold. It seems, based on what I can find, that it does tie into the CS portion of CCA. I'm going to keep digging to see if I can find stuff expressly confirming this.Anonymoushttps://www.blogger.com/profile/09214344597064983852noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-91660161543179169282014-08-17T14:30:17.697-05:002014-08-17T14:30:17.697-05:00RX-SOP is not meant to asses the medium, it's ...RX-SOP is not meant to asses the medium, it's really another tool to reduce cell size, like trimming data rates and limiting power level. The practical application to RX-SOP is to keep clients on the AP/Cell they are suppose to be on. So for example in a stadium, you wouldn't want clients sitting in a lower bowl, attaching to an AP in the upper bowl. So setting the RX-SOP at a value that will ignore packets with a certain RSSI value from being processed, makes the client search for the better AP. From an AP perspective, the CU does go down slightly, because it thinks there are less 802.11 frames on the medium. From a practical perspective, the CU doesn't really go down because the transmissions are still there. In order to address CCI you need to change the CCA value, so that the AP can transmit at values higher than -85dBm.Anonymoushttps://www.blogger.com/profile/09214344597064983852noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-33722629376299518722014-08-17T12:48:23.044-05:002014-08-17T12:48:23.044-05:00Good point; so if CCA-CS stays the same, and all w...Good point; so if CCA-CS stays the same, and all we are really doing is not sending an ACK, that doesn't help with determining the medium as free. It would not help with the AP's perception of CU either, assuming the client moved to a different AP on same channel. You mention it does not change the value the AP uses to determine when & when not to transmit, but you also mention that any packet weaker than SOP threshold is "ignored/not processes". I guess the question is "how much" ignoring does the AP do? If it doesn't ignore enough to determine medium "free" (ie only implements a "no ACK" as you said), that would be contrary to the notion of treating the frame as noise,tuning sensitivity, ear-muffs etc. Pg. 90/91 of BRKEWN-3010 for example.Anonymoushttps://www.blogger.com/profile/12623366983649605110noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-36020061295141160582014-08-17T11:50:09.303-05:002014-08-17T11:50:09.303-05:00Travis, your comment is interesting. Does this mea...Travis, your comment is interesting. Does this mean that even WITH RX-SOP implemented that a Cisco AP will still show CCA as busy when neighboring AP or client frames are below the RX-SOP threshold but above -85dBm (or whatever Rx Sensitivity allows proper preamble and PLCP decoding)? So RX-SOP doesn't necessarily help alleviate CCI in an HD environment if that is true. Andrew von Nagyhttps://www.blogger.com/profile/12658799453646609565noreply@blogger.comtag:blogger.com,1999:blog-1988432060681510848.post-57856651817667416682014-08-17T10:32:22.010-05:002014-08-17T10:32:22.010-05:00Just to clarify. RX-SOP doesn't change the CC...Just to clarify. RX-SOP doesn't change the CCA value the AP uses to determine when and when not to transmit. It uses the value of incoming packets from clients, to determine if it ACKs the packet. So if RX-SOP is set at -78dBm then any packet the AP receives at -79dBm or worse, it ignores and does not process. The CCA value of the AP, which is -85 by default stays the same after RX-SOP tuning.Anonymoushttps://www.blogger.com/profile/09214344597064983852noreply@blogger.com