Jump to content

boomerbubba

S4GRU Premier Sponsor
  • Posts

    704
  • Joined

  • Last visited

Posts posted by boomerbubba

  1. Thinking about the sector ID's (and locations) that Sprint towers seem to employ, I got to wondering a few things.

     

    What is the advantage for Sprint of using sector ID's off of a given tower instead of just using the tower ID itself? Troubleshooting?

     

    Does the location of the sector correlate to the "aiming point" for the given towers downtilt, or is it just an arbitrary location in the direction the sector faces?

     

    The site near my house has recently stopped reporting location (at least on the sector that serves me). It now reports as unknown. It's been in this state for the last 36+ hours. Does this indicate anything? While I haven't seen roaming indicated in the status bar, my S3 is set up to show dual clocks on the lock screen when I am roaming and it's displayed this multiple times since the site stopped reporting sector locations. Ideas why?

     

    I know it's multiple questions in a single thread, but I'm a multiple questions kinda guy.

     

    As for why they are used, that much is pretty obvious. The "tower" is just a location. Your connection is actually to one of the sector radios on each site.

     

    I assume you are talking about CDMA sectors, since they broadcast coordinates and LTE sectors do not. It is not a question of "advantage" for Sprint in using them. These sector IDs (Base Station IDs) are baked in the cake of the CDMA standard. Generally, the devices' own telephony is based on sectors. For some other purposes, Sprint does use its own site IDs. Presumably, Sprint's internal databases link from the more granular sector IDs to their own tower IDs comprehensively. So when troubleshooting based on sector, they can always know the tower.

     

    As for your location question, be aware that some Sprint towers' sector radios do broadcast the actual tower location's coordinates, identical for each sector, while other towers broadcast a different, offset coordinate for each sector. I do not know why this is, but it is. In my own experience, it does seem that the size of the offset is proportional to the designed coverage size of each sector.

     

    This subject is summarized in this FAQ comment, which contains a link to a fuller Sponsor thread:

  2. Net Monitor showed me connected to the tower at Carmel and Pineville-Matthews Rd with LTE. I haven't seen LTE from this tower before.

     

    Netmonitor does not identify the site of LTE connections, only CDMA connections that happen to occur while you have an LTE connection from somewhere. That might not be the same place.

    • Like 1
  3. Please don't take this wrong, I'm not criticizing, I'm missing something. Why not just run Sensorly? Does it discover LTE better? My plan was to just run Sensorly when LTE pops up in St Louis and let it plot where its at on their maps. Does toggling airplane mode help that much on a phone like the Note 2 that doesn't have the issues the LTE EVO users report?

     

    Sensorly doesn't do anything to "discover" LTE. It depends on the phone itself to make the connection, and just logs whatever connections are present. Running Sensorly or any other app that is sending and receiving data might even interfere with discovering LTE, because the phone may avoid switching to 4G while it has an active 3G data connection in progress. The thing that forces a scan, which then might make a 4G connection if a signal is present, is the workaround of toggling airplane mode.

    • Like 1
  4. My own wish is for an app that would log my LTE connections so I can correlate them to towers offline after the survey: The log should include datetime stamp, user lat/lon coordinates, signal RSRP, signal RSRQ, Sector Physical Cell ID integer, and Sector Cell Identity (a 28-bit value) as a hex string.

     

    As it stands now, these two forms of LTE sector IDs are available manually on the LTE Engineering screens of different devices. But Samsung devices show only the Physical Cell ID; EVOs show only the Cell Identity values. We need both, so the same LTE tower sites must be surveyed twice -- once by each type of device. An app like this can do it all at once. Boths sets of LTE IDs are available in the Android telephony API for Jellybean, but not earlier Android versions..

    • Like 1
  5. I just saw the completed sites map. I see you use the Sprint site codes. The Douglas site - - - edited - - - and antennas are up -- not sure if it is testing yet. New Hope site - - - edited - - - just went in so I'm surprised Sensorly already showing signal -- I'm pretty sure that is testing. Hwy 36 and Rice is site - - - edited - - - and I'm surprised that is not showing any signal yet -- antennas have been there a couple weeks -- but again probably testing.

     

    Others which have been up for months Byron site Sprint-NPCS site - - - edited - - - , and Owatonna site - - - edited - - - .

     

    There is a reason that the NV Sites Complete map is in the Sponsor-level Interactive S4GRU Maps forum: The content is for Sponsors. Republishing such detail in this public forum violates S4GRU rules. The Sponsor forums provide a venue to discuss such details.

  6. Robert has an article on the Wall about Sprint's latest announced list of cities, to which he has inserted the following comment (boldface emphasis in his original).

     

    There seems to be a lot of confusion out there, especially among Facebook readers, that cities have been removed from Sprint's LTE deployment list. THIS IS NOT THE CASE! The list above are "additional" cities being added to the list. Sprint has now announced approximately 150 cities total where work is under way. No cities have been removed. In fact, Sprint will start work in every market in 2013. Sprint is deploying LTE nationwide as a part of its Network Vision upgrades.
  7. How do you know it's a Sprint tower?

     

    It was being set in place with a previous sprint tower, and i used open signal maps to find it

     

    This begs the question of how you knew the "previous Sprint tower" was actually a Sprint tower. And what does "being set in place" with it mean? Sounds like there are multiple towers at this location, and Sprint may or may not be sited there.

     

    Also, Open Signal is notoriously wrong about towers. It does not even purport to map actual tower locations, only imputed geographic coordinates derived -- usually inaccurately -- from crowdsourced signal-strength collection.

  8. Earlier today I had an LTE RSRP of -82 but only got 12.3 down and 7.9 up. I mapped the base station (CDMA Field Test) and it was incredibly close to where I was. Now 12.3 down is certainly fast enough for whatever I would need, but it's pretty depressing that being that close with that strong of a signal is now only getting 12.3 down. Shouldn't that signal be pulling in high 20s? There have been many times where I've had RSRP of -105 to 115 and I'm getting 2-3 down.

     

    Is this an indication that as more people start using the Sprint LTE network that the speeds will suffer?

     

    The mapping of the CDMA base station is unrelated to the source of your LTE signal, which may be coming from a different site altogether. Further, for unrelated reasons, even the mapped CDMA coordinates may not reflect the actual CDMA site because some base stations broadcast CDMA coordinates that are far offset from there.

     

    So your mapping experience is irrelevant. More important, you empirically logged a pretty strong LTE signal and moderately fast speeds. That could just be load related.

  9. downloaded cdma field test, now the primary tower is a totally different location. wrong again however, only single story residence, no towers, poles, tree with panels (really by the way?).

    My head actually cocked to the side when I read that as well! Aside from the issues with anchoring panels to something that is alive and potentially growing, I couldn’t figure out how something like downtilt would work on something that has the rigidity of a real tree. Can you imagine what your service would be like in a storm? “I have signal! No, wait neverm… THERE! Oh, gone again.” It’d be like a giant metronome.

     

    http://waynesword.pa...du/faketree.htm

     

    These fakes are pretty spot-on from a distance! This is not good news for my driving skills since S4GRU.com has taught me to keep an eye out only for conspicuous panels. Now I'm going to start questioning everything.

     

    A red herring.

     

    It is obvious that danny1st was not looking at the actual tower site, but was being fooled by the offset coordinates being squawked by the CDMA sector base stations. When the coordinates are offset from the actual antenna site, they might be far way from there. Even miles away in some cases.

     

    Yes, there are "stealth" towers disguised as trees, etc. But that was not the case here. This was a case of the user not understanding that the mapped coordinates displayed in the app might not be real.

     

    BTW, since you have Sponsor privileges, you might also be interested in this thread: CDMA towers that squawk the wrong coordinates

  10. What do you think would be as close as possible to what I had on my blackberry signal loc app you would say to get?

     

    Either CDMA Field Test or Netmonitor will export KML files for the squawked CDMA coordinates, warts and all. I prefer CDMA Field Test. (I used to use Netmonitor, but I uninstalled it because I decided I didn't like its privacy risk. Others may weigh that risk differently than I do.) Netmonitor did show multiple sites on its interactive screen. Of course, the log files for both include multiple sites. Both apps export KML files plotting the coordinates broadcast by CDMA sector base stations the handset connects to, which may or may not be the actual antenna sites. The output is still useful if you understand these limitations, and have access to the S4GRU master maps of the actual Sprint sites for reference to divine the actual site within a plotted triangle.

    • Like 1
  11. Wish someone would come up with an app like I had for my Blackberry phone before going over to the Note 2 called Signal Loc. That signal loc app for the blackberry phone was a cool app that helped me locate the broadcasting Sprint tower and even roaming tower locations and then saved it to my SD card in a file and then exported it out to google maps to see a wider range of where I traveled and where all the towers it found. If only someone could create an app for the Android OS system for people like me that love going out and finding cell towers like that I would jump right on it and download that app for my phone here.

     

    CDMA FIeld Test and Netmonitor will export logs as CSV files and/or as KML files for mapping. But the content of the coordinates mapped has all the shortcomings of the interactive displays I described above, because those shortcomings originate with the Sprint base stations themselves. So I am sure your Blackberry app had the same limitations.

    • Like 1
  12. I understand and thanks for the response. just sad that we as consumers have to seek information on secondary third party sites and that their information is misleading or just outright false. if they are upgrading towers, you would think it would be in there own interest to give out valid information.

     

    on that note, signal finder and open signal are indicating that the tower I am currently connected to is at site 1 which is actually a tower (mono pole) and cdma field test is within a few hundred feet from this tower that is not identified as a Sprint tower according to their own information (stay with me here cause I now your ripping your hair out... I do get what your saying). but the tower that is getting the updates that is a few hundred feet closer to my location, I am not connecting to according to all these signal apps. the apps are showing actual tower locations for this tower...

     

    I guess it's not that important really, just hearing about all these upgrades, and seeing Sprint proudly release all the improvements on their towers and yet absolutely zero improvements on actual speed. just frustrating.

     

    You really cannot generalize about all purported "tower mapping" apps because they work differently The apps you are using (Open Signal and Signal Finder) do not show actual tower locations at all, only imputed locations that they have computed using crowdsourced signal strength collections. These computations are extremely and notoriously unreliable. I consider these apps, when used for this purpose, to be junk.

     

    At least the other class of apps (CDMA Field Test, Netmonitor, etc.) plots the locations that are broadcast by the CDMA base stations. In some cases, these locations are the actual sites so it is straightforward to see them. But in others, the locations do not represent the sites, but rather the offset locations, as the FAQ comment explains.

     

    Yes, it would be nice if Sprint and other carriers would just publish their full set of antenna sites, but they don't. For Sprint, the single best resource is the set of Sponsor maps here at S4GRU, which are sourced from internal Sprint project documents. With these Sponsor maps, CDMA Field Test and a little work, it is possible to isolate the CDMA site you are connected to. I have always been able to do it, but no app will do all the work for me. (BTW, the other major CDMA carrier, Verizon, goes even further to suppress tower locations. Even though Verizon sites apparently broadcast coordinates, the carrier seems to cripple the Android API on its handsets to block apps such as CDMA Field Test from mapping the sites at all.)

     

    None of this applies to LTE, which does not even include broadcast coordinates in its technical standard. So even with the S4GRU maps, the only way to correlate their mapped LTE sites to signal data available on the handsets is by field survey to build our own lookup table. No doubt Sprint has such tables internally, and it would be nice if they were published. But they are not.

  13. downloaded cdma field test, now the primary tower is a totally different location. wrong again however, only single story residence, no towers, poles, tree with panels (really by the way?). just want be sure I am connecting to the Sprint tower that is receiving the updates and if not, why? I know Sprints data is very poor in some locations. but when I have a tower within 500 feet with no obstructions (my road leads directly to it as if it was the fricken yellow brick road) and that tower, according to Sprint has received multiple speed and data capacity upgrades in the past year... why I'm I still pulling less than 100 k down during off peek hours.

     

    Sorry for the b#&$h fest, just trying understand?

     

    CDMA Field Test is mapping the coordinates broadcast by the base station site (tower). But those coordinates may not be the tower's actual location. The coordinates broadcast for the sites' sectors may be offset from there. That is why I said above, "You still might have to do some field work to figure out the exact sites." And you would probably need the S4GRU maps of all the Sprint towers, available to Sponsors here, to do that.

     

    See this comment in the FAQ thread for further explanation.

  14. That is just stupid that you would write to a file the 3G tower your connected to in relation to the 4G connection. What good what that file do? On the Evo, when connected 4G looking at LTE Engieering, the serving cell site is 04c11302, the 3G cell site that it is connected to is nid: 41 sid: 4139 bsid: 1826.

     

    I have seen where most of the serving sites I am around in Overland Park start with 04c0 or 04c1.

     

    Congratulations. You now have taken the initial step toward mapping LTE sites by their actual LTE IDs. But these have nothing to do with the NID, SID and BSID values, which all pertain to the CDMA hierarchy. Eventually these two sets of radios (CDMA and LTE) will tend to coexist on the same Sprint tower sites, but the only way for us to correlate them is by empirical surveys. Presumably Sprint has internal databases that can relate them, but we don't have access to them. Neither do our handsets, AFAIK.

     

    We think the LTE Record is just a crude way to build a local set of CDMA IDs that might trigger an optimized scan for some LTE signal. It might be good for that, but nothing else I can think of. It does not correlate specific LTE IDs to specific LTE IDs.

  15. Why does the LTE record include SID, NID and BSID if LTE does not squak them? Or, if LTE record can get them, why can't the mapping apps?

     

    I think the consensus speculation is that the LTE Record is just a list of 1x CDMA sites to which the device happened to be connected while it had an LTE connection. These are not necessarily the same sites as the one that had the LTE signal, and in a thinly developed market probably are not the same. These CDMA sites likely don't even have LTE upgrades yet.

     

    As best as anyone can tell, this "LTE Record" might form the basis for some scheme to optimize scanning for LTE sites. No one knows for sure. But we do know that mapping them would be a nonsensical way to show LTE. It would be pure misinformation.

     

    Some of us at S4GRU do know how to map the real LTE sector IDs using the LTE Engineering screens of most devices. We have reverse-engineered that much. But this takes a lot of work by users doing field surveys to capture the IDs and triangulate them so they can be correlated to antenna sites. It is not an automated process by any means, and no mapping app will do it all for you. To my knowledge, no one is even trying to map LTE sites by their sector IDs, except for those of us in the sponsor thread for Austin and nearby areas. We have mapped almost every live tower by its LTE sector IDs.

    • Like 1
  16. If you are trying to track you LTE connections to specific antenna sites, there is no app that will do that for you directly. You have to survey the towers/sites yourself using your LTE Engineering screen.

     

    If you are trying to track CDMA connections, the apps you are using won't help either. Use CDMA Field Test. Or if you don't mind the privacy risk, try Netmonitor. You still might have to do some field work to figure out the exact sites.

     

    In any case, start with a comprehensive map of the Sprint towers, which you can get as an S4GRU sponsor. Sprint's own maps are incomplete.

    • Like 1
  17. Sensorly shows the center of the recorded LTE in the same spot as the tower.

     

    Well, that Sensorly hotspot is certainly evidence of an LTE tower. It may not be conclusive evidence that it is your particular tower, especially if there are any other possible Sprint towers in range, but it is evidence that poses the hypothesis to prove or disprove. (I wish you had mentioned that Sensorly fact before. The Speedtest results by themselves don't show tower location.) And it really does help to see the Sponsor maps because they show where the Sprint sites are.

     

    The best way to know for sure is to go close to the tower site and log the RSRP signal strength all around it using the LTE Engineering screen on your phone, seeking strong signals for all sectors. You also can log the values for each sector's Serving Cell ID. Then you can replicate the test from your original position to see if the ID matches.

×
×
  • Create New...