Menu

#36 wrong Cell ID

open-later
Nick
5
2012-01-02
2012-01-02
Anonymous
No

openbmap on Android logs wrong Cell ID (I think it's only on 3G, haven't checked with 2g) for example cell:
247 02 115 7638 is logged as 247 02 115 1386262
8418 as 1384674

Discussion

  • Nick

    Nick - 2012-01-02

    not sure to understand where the problem comes from.

    To draw the map, I apply a 0xFFFF and operation to keep the cell id inferior to 65535 as mentioned in GSM specification (and by experience of 3G, 3G+ cell phones)

    So 1386262 should appear as cid 10006 on the map
    1384674 should appear as cid 8418 on the map

     
  • Anonymous

    Anonymous - 2012-01-02

    It seems that in mentioned cid's are correctly displayed on the map, but in different region map displays 13... cid's, almost all 2G cells (not 2G+) have wrong names:
    they are shown on map, but no view zone is available
    http://www.openbmap.org//with_osm5.php5?mcc=247&mnc=2&lac=115&cellid=1385244&zoom=14

     
  • Nick

    Nick - 2012-01-02

    If the map displays cid greater than 65535, it means the map generation went wrong.
    is it the problem you are refering to ?

    if yes, I will generate the map for Latvia from scratch.

     
  • Anonymous

    Anonymous - 2012-01-02

    Please regenerate it from scratch, I will inform if I encounter this issue in the future.
    Many thanks!

     
  • Nick

    Nick - 2012-01-02

    I ll do that in the next days.
    thanks for posting the problem.

     
  • Nick

    Nick - 2012-01-02
    • status: open --> open-later
     
  • Nobody/Anonymous

    "To draw the map, I apply a 0xFFFF and operation to keep the cell id inferior to 65535 as mentioned in GSM specification"
    I'm not sure I understand the reason for this.
    I constantly log cells with cellid number higher than 65535 and I think it's quite confusing to see them on the map renamed with a different number.
    Moreover I see that some cells (maybe imported from another database ?) are shown with the actual cellid (even if with a wrong generation, i.e. 2g instead of 3g), e.g. 222 99 40045 1603227

     
  • Nick

    Nick - 2012-01-05

    The map generation went wrong also for Italy and probably others. I ll generate the maps on request. Please note that the offline data are not wrong and cid are inferior to 65535 in sqlite database.

    We can now decide to display a cid inferior to 65535 at the logger interface as it is less confusing.

     
  • Nobody/Anonymous

    > We can now decide to display a cid inferior to 65535 at the logger
    > interface as it is less confusing.

    Or (better, under my point of view) store the cid in the database and show on the map the cid numbers as they are recorded, i.e. above 65k.
    I understand what you say about the GSM specifications, but all the netowork monitoring tools I've installed on my android phone (Cell connectivity tracker, Network info II, OpenCellId Client, Umts status) return the same above-65k number, and in my opinion it would be wise to keep your project coherent with the others. This would erase any source of confusion.

    My two cents.

     
  • Nick

    Nick - 2012-01-07

    All right.
    We would need a new view/map to visualize raw data (with cid > 65535).
    We will also keep the processed data view/map.(with cid < 65535)

    The openBmap logger clients will still display the collected raw data (cid > 65535)

    For processed data maps wrongly displaying cid > 65535, we will generate the corrected maps on demand.

     
  • Anonymous

    Anonymous - 2012-01-09

    1. Not all the network monitoring tools on Android show CID above 65k, my favs Netmonitor and G-Mon show CID correctly, that was the reason how I discovered that OpenBmap client showed wrong CID.
    2. Can you make that CID would be displayed in 2 ways? like: 8418 (1384674)
    3. In Latvia on regenerated map cells with wrong IDs have disappeared, but some correct cells' markers are missing on map: http://www.openbmap.org//with_osm5.php5?mcc=247&mnc=2&lac=190&cellid=8412&zoom=15

     
  • Nick

    Nick - 2012-01-10

    OK with point 2.

    I ll check point 3 Asap.

     
  • Nick

    Nick - 2012-01-18

    If you agree I am first investigating the galaxy S2 bug then I check the problem with the "correct cells' markers missing on map".
    The problem is found and exists and it needs an investigation.

     
  • Nick

    Nick - 2012-01-28

    I have created a bug on the backend server side about duplicated cells.

    I am still investigating the markers missing.

     
  • Nick

    Nick - 2012-01-28

    I have found the problem with the markers missing.
    I have generated Latvia and Italy again.

    It should be find now. please tell me if it is ok.

     
  • Anonymous

    Anonymous - 2012-01-28

    Thank you for fast response and solving this issue! But there is another one :)
    The reason why I noticed duplicated cells was this particular cell ID:
    http://www.openbmap.org//with_osm6.php5?mcc=247&mnc=2&lac=180&cellid=5141&zoom=13
    I was driving in the area of current marker location to the west and my phone was receiving good signal strength so I can confirm that approximate marker location is in right place. Later I wanted to check it’s coverage zone and found it in other place. Now after regeneration old and wrong marker has disappeared but the coverage zone is in the old position and the new zone is not regenerated.

     
  • Anonymous

    Anonymous - 2012-01-28

    Sorry my mistake! Now the zone is regenerated correctly!

     
  • Nick

    Nick - 2012-01-29

    I love good news.

    So this bug is now related to
    "CID would be displayed in 2 ways? like: 8418 (1384674)"
    only

    thanks!

     

Log in to post a comment.

MongoDB Logo MongoDB