maemo.org - Talk

maemo.org - Talk (https://talk.maemo.org/index.php)
-   Applications (https://talk.maemo.org/forumdisplay.php?f=41)
-   -   [Announce] Cellnet-Info for N900 (https://talk.maemo.org/showthread.php?t=89238)

ade 2013-02-20 16:01

[Announce] Cellnet-Info for N900
 
3 Attachment(s)
Netmon like program to provide cellular info. Reason for creation was to learn a bit more about cellular networks and dbus, and I like apps to support portrait mode :)

What does it offer:
  • SIM related info
  • Current cellular info
  • Scan of available mobile networks
  • Find country- and operator codes
  • Landscape/portrait support
  • Help pages for explaining the related terms (mainly based on wikipedia info)
  • Display your approx. postal location using celltower info (internet connection required) using Google API/openCellId

The package can be found in the extras-devel repository.

P.s.:
- The country and operator codes are retrieved from an internal list and will not be up-to-date nor complete.

Tiran 2013-02-20 17:11

Re: [Announce] Cellnet-Info for N900
 
In an installation time dependence python-pyside.qtmaemo5 demands. ;)

sixwheeledbeast 2013-02-20 17:43

Re: [Announce] Cellnet-Info for N900
 
Very nice. IMO we needed a replacement for netmon as it was crashing for me intermittantly.

praveenchand 2013-02-21 06:48

Re: [Announce] Cellnet-Info for N900
 
Nice application, is there any way to get RF parameters like RSCP, EcNo etc etc?

Estel 2013-02-21 08:32

Re: [Announce] Cellnet-Info for N900
 
Instant subscribe, download and testing :) No bugs so far, despite pestering it with modem turned off, or during mode change - got "ade's quality" stamp on it, good work and thanks.

Would it be possible to measure signal strength of mobile networks in area, when querying them via "Mobile Networks" tab? It would become very handy tool to probe area for best GPRS/3G data plan/prepaid, instead of buying one prepaid per every operator, and testing signal one-by-one.

Also - but you probably thought about it yourself, already - maybe some switch to trigger auto-update of data, at user configurable interval? (1 sec, or 2 sec, or 3 sec, or...).

/Estel

misiak 2013-02-21 09:44

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by sixwheeledbeast (Post 1324130)
Very nice. IMO we needed a replacement for netmon as it was crashing for me intermittantly.

Netmon always crashes for me when I'm in roaming (not only international, but also inside my own country, when N900's operator widget reports different network name - my operator apparently has a deal with second one for areas with low or no signal at all, as even the old 3310 behaves like that). I'll try this app in this scenario, soon:)

sixwheeledbeast 2013-02-21 16:42

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by misiak (Post 1324221)
Netmon always crashes for me when I'm in roaming

IIRC there is an error if run in Terminal, when Netmon crashes.
I have never remembered to report it.

ade 2013-02-21 20:53

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by praveenchand (Post 1324203)
Nice application, is there any way to get RF parameters like RSCP, EcNo etc etc?

First of all I am not a cellular network expert, so I don't know the exact differences with the current signal indication.
Secondly there is very little official N900 documentation on these matters. Most is found out by the Maemo community itself.
I doubt it if this info can be retrieved, but if you can point me to some commands....

Quote:

Originally Posted by Estel
Would it be possible to measure signal strength of mobile networks in area, when querying them via "Mobile Networks" tab? It would become very handy tool to probe area for best GPRS/3G data plan/prepaid, instead of buying one prepaid per every operator, and testing signal one-by-one.

Would be nice indeed, but in this case the question is also: how to retrieve this info?

Quote:

Originally Posted by Estel
Also - but you probably thought about it yourself, already - maybe some switch to trigger auto-update of data, at user configurable interval? (1 sec, or 2 sec, or 3 sec, or...).

That's why I explicitly mentioned the info is static. If I would make it dynamic, I would update it by changing values. I have been playing with that, but abandoned it when I couldn't get it working properly. Maybe I'll try a second time in the future.

Quote:

Originally Posted by misiak
Netmon always crashes for me when I'm in roaming (not only international, but also inside my own country, when N900's operator widget reports different network name - my operator apparently has a deal with second one for areas with low or no signal at all, as even the old 3310 behaves like that). I'll try this app in this scenario, soon

I did not have the opportunity to test it with any other operator than my default one, so please do test :)

P.s. QtMobility should be able to give a little more info (like if a dual sim card is used), but with this module loaded, PySide crashes everytime in my environment.

peterleinchen 2013-02-21 21:57

Re: [Announce] Cellnet-Info for N900
 
Nice one. Good one!
It does not offer more information on network than netmon, right?
But SIM info is pretty cool.

Estel's wish to have info on scanned networks would be very handy. But I also do not know it is possible at all.

One remark regarding the scan:
Could you please add a 'Cancel' button? It is annoying to have no chance to abort the scan in any way.

ade 2013-02-21 22:21

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by peterleinchen (Post 1324329)
Nice one. Good one!
It does not offer more information on network than netmon, right?
But SIM info is pretty cool.

Estel's wish to have info on scanned networks would be very handy. But I also do not know it is possible at all.

One remark regarding the scan:
Could you please add a 'Cancel' button? It is annoying to have no chance to abort the scan in any way.

No, apart from the country name displayed if known and the option to lookup MCC/MNC codes, the network part has no extra info.

I removed the cancel button (which is default), but I will see if I can make it work gracefully once I put it back.

mr_pingu 2013-02-21 22:47

Re: [Announce] Cellnet-Info for N900
 
I agree with Estel_ it would be nice to have something like to probe all networkcells and display info about them. Sort of promiscius mode/kismet like for celltowers only.

However I am well aware theres many closed / not documented stuff on this subject, too bad :( Anyway thanks for this program, Ade :)

Also, it reminds me of this application for s60/symbian:
http://www.patrickfrei.ch/phonetinfo/index.html

ade 2013-02-24 10:37

Re: [Announce] Cellnet-Info for N900
 
New version in the opening post, which deals with improving the progress bar while scanning networks.

If I am going to work on it more, refreshing to displayed information will be the focus.

P.s. some more screenshots (and info if you can read russian :)) can be found here: http://linuxcenter.kz/page/cellnet-i...emo-nokia-n900

Tiran 2013-02-24 17:13

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1324759)
P.s. some more screenshots (and info if you can read russian :)) can be found here: http://linuxcenter.kz/page/cellnet-i...emo-nokia-n900

It I on linuxcenter.kz have made news to russian-speaking users (I hope you not against?). :D

ade 2013-02-28 18:50

Re: [Announce] Cellnet-Info for N900
 
Version 0.0.1-2 now to be found in the openingspost.

This time most values displayed will be updated when they change (signal strength, operator, cell-id etc).

As soon as the display is off, changed signal receiving is turned off. This could mean a little battery saving, but also a not 100% accurate state display just after the screen is turned on again.

Going through the sources of Netmon was very helpful for me, so thanks to Spag for creating this program.

Wikiwide 2013-02-28 22:13

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1326026)
As soon as the display is off, changed signal receiving is turned off. This could mean a little battery saving, but also a not 100% accurate state display just after the screen is turned on again.

Quick reply...
Good. It's the same with the time displayed in hildon-status-bar: it doesn't update itself when the screen is off.
Best wishes.

ade 2013-03-16 00:00

Re: [Announce] Cellnet-Info for N900
 
As the autobuilder is working again: the latest version is now available in extras-devel.

nokiabot 2013-03-16 01:00

Re: [Announce] Cellnet-Info for N900
 
Can it be used to automate the n900? Lyk invoking actions when it enters specific cells:)

Estel 2013-03-16 01:55

Re: [Announce] Cellnet-Info for N900
 
We have other package for that (actions, when it enters specific cell[s])

ade 2013-06-03 16:58

Re: [Announce] Cellnet-Info for N900
 
Version 0.0.1-3 is send to extras-devel.

A short explanation on the latest change:
Choosing "Current cellular info" will show extra information if an internet connection is present. It will search and display the geographic coordinates of the celltower found on opencellid.org. These coordinates are then used to find and display the postal location of the celltower (using maps.googleapis.com data). The info should be updated if the cellid changes.

I don't know how thorough the opencellid.org databases is, but it is way to big to store locally on the phone :)

peterleinchen 2013-06-04 09:29

Re: [Announce] Cellnet-Info for N900
 
Hey ade,
thanks for that little nice feature.
Not yet updated, so question in advance: is it possible to select/deselect that internet access in settings?
If not, I would like to ask for this? (I do not like to spread my coordinates througout the internet without knowing)

pichlo 2013-06-04 09:44

Re: [Announce] Cellnet-Info for N900
 
@peterleinchen, you are spreading your coordinates anyway (through your IP address - granted not as accurate as cell tower location).
If you are concerned, you can always go offline before running cellnet-info.

peterleinchen 2013-06-04 13:18

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by pichlo (Post 1349422)
@peterleinchen, you are spreading your coordinates anyway (through your IP address - granted not as accurate as cell tower location).

I know, I know (proxy, tor, anonymous, ...)!
Quote:

Originally Posted by pichlo (Post 1349422)
If you are concerned, you can always go offline before running cellnet-info.

I just do not like programs that are shouting out to the internet.(when I am not offline).

Estel 2013-06-04 17:14

Re: [Announce] Cellnet-Info for N900
 
Well, simple "toggle" in options to enable or disable enhanced info (*needs internet connection) wouldn't hurt.

ade 2013-06-04 18:40

Re: [Announce] Cellnet-Info for N900
 
New version (0.0.1-4) made searching for cell tower location optional.

pichlo 2013-06-04 19:04

Re: [Announce] Cellnet-Info for N900
 
Well, in my case the cell tower location showed up a whole 8 minutes off.
Still, thanks for the update ;)

peterleinchen 2013-06-08 04:40

Re: [Announce] Cellnet-Info for N900
 
Hey ade,

I found a small bug/glitch.
When you have a network connection, but that network does not provide full internet access (like behind the "Great Firewall"), then after trying to fetch location info, the program is frozen and needs a restart.

Thanks.

ade 2013-06-16 22:02

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by peterleinchen (Post 1350429)
Hey ade,

I found a small bug/glitch.
When you have a network connection, but that network does not provide full internet access (like behind the "Great Firewall"), then after trying to fetch location info, the program is frozen and needs a restart.

Thanks.

Peterleinchen,
I now tried to include a 10 seconds timeout.
As the autobuilder is stuck frequently (version 0.0.1-6 which fixes a utf-8 issue is still waiting there), could you try cellnet-info_0.0.1-7 for me?
Thanks in advance.

peterleinchen 2013-06-17 13:28

Re: [Announce] Cellnet-Info for N900
 
Hey ade,
sorry to say that this does not fix the behaviour.

I tried it with 1-6 (which I was informed about to update a week ago) at home, but could not reproduce. So I reinstalled to 1-5, but could also not reproduce (tinkering with local ad-hoc WLAN and my hosts file).
Either it worked or I got a pop up.

But today I passed a WiFi Hotspot and here I could reproduce it.
After pushing "current cellular info" the app is still responding to button pushes, but no action happens anymore. Seems like a endless loop/timeout in background.
So my guess is you check via nslookup (or similar) for resolving of hostname and if successful, then try to fetch info? This would explain behaviour I have seen, as this hotspot needs some http site access/registration before you have real internet access. Would like to check myself in sources, but I am a bit handycapped atm.

ade 2013-06-17 14:17

Re: [Announce] Cellnet-Info for N900
 
For the record: you did test with cellnet-info version 0.0.1-7 (found in the attachment)? Because that's the one I adapted for this, and you are mentioning reinstalling version 0.0.1-5..

pichlo 2013-06-17 16:01

Re: [Announce] Cellnet-Info for N900
 
Sorry to barge in like this but don't all net-aware apps suffer the same thing at public hotspots with a login portal? That's at least my experience: email doesn't update. navigation apps can't find themselves, Skype can't log on... until you go to the damn portal and type your creds.

FWIW, I have installed 0.0.1-7 from the attachment and it works fine but I have yet to try it at a local supermarket with such a hotspot.

BTW all versions consistently show the cell tower location about 8" (about 8 miles or 13 km) off. That's of course a database error, not yours but I wonder how do such things happen? My provider moved a tower and forgot to update the database?

ade 2013-06-17 16:48

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by pichlo (Post 1352715)
BTW all versions consistently show the cell tower location about 8" (about 8 miles or 13 km) off. That's of course a database error, not yours but I wonder how do such things happen? My provider moved a tower and forgot to update the database?

Not entirely my experience. Sometimes it is a location about 7km away, sometimes about 2km away. A few weeks ago it pointed exactly to a large building I was visiting. Used cellid's do change every now and then, even if you are not on the move.

If you really think your cell tower coordinates are incorrect, you could report it at http://www.opencellid.org. Providers keep there celltower locations secret, the information is purely bases on community input.

peterleinchen 2013-06-17 18:23

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1352704)
For the record: you did test with cellnet-info version 0.0.1-7 (found in the attachment)? Because that's the one I adapted for this, and you are mentioning reinstalling version 0.0.1-5..

Yep. Sorry for being unclear on this point.
I first tried with 1-6, then reimstalled 1-5 and then of course tested 1-7.
@pichlo Yes, a lot of those hotspots work like that. And that is where you may see this behaviour (also/first had it behind a hotel hotspot WITH credentials typed, but China).

@ade Sorry, didnt have time to test more, had to catch my train.

--edit
Checked one more time with 1-7 and it indeed returns. But only after more than 30 seconds. That long time no response at all.

Estel 2013-12-10 11:41

Re: [Announce] Cellnet-Info for N900
 
I was wondering, if this:
http://talk.maemo.org/showpost.php?p...0&postcount=30

...could be (ab)used to get information about band we're using and show it in cellnet?
---

Also, "kinda" request - currently, "signal strength" is funny, I get 100% all the time, despite ~10 dbm differencies. Would it be more sane to calculate it basing on CSQ (which is based on SNR), where CSQ 32 = 100% signal (impossible in real-life), and CSQ 0 = 0% (no signal)? It would give some meaningful results, as current signal strength % is quite useless.

Also, just a dumb question - shouldn't it be -n dBm, instead of n dBm? Aren't RSSI ratios presented as negative values (higher = closer to 0 = better)?

/Estel

ade 2013-12-10 20:12

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by Estel (Post 1394471)
I was wondering, if this:
http://talk.maemo.org/showpost.php?p...0&postcount=30

...could be (ab)used to get information about band we're using and show it in cellnet?
---

Although it refers to Phone.Net, I have no idea how to actually query that info. If someone can, I am more than happy to apply it.

Quote:

Also, "kinda" request - currently, "signal strength" is funny, I get 100% all the time, despite ~10 dbm differencies. Would it be more sane to calculate it basing on CSQ (which is based on SNR), where CSQ 32 = 100% signal (impossible in real-life), and CSQ 0 = 0% (no signal)? It would give some meaningful results, as current signal strength % is quite useless.
I do not agree on the useless part. It's equal to the bar value in the status bar as far as I can tell. You must be blessed with a very good reception, as my signal strength varies between 100% and no signal at all from time to time. I doubt if "at+csq" results in a significant other pattern, but I will do some more measurements on that.

Quote:

Also, just a dumb question - shouldn't it be -n dBm, instead of n dBm? Aren't RSSI ratios presented as negative values (higher = closer to 0 = better)?
/Estel
I guess you are right about that, I will fix that in a next release. And I will try to add a little more cellular info.

misiak 2013-12-10 20:39

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1394640)
Although it refers to Phone.Net, I have now idea how to actually query that info. If someone can, I am more than happy to apply it.

In terminal, try
Code:

dbus-send --system --print-reply --type=method_call --dest=com.nokia.phone.net /com/nokia/phone/net Phone.Net.get_current_cell_info
for other functions of "Phone.Net" object and what the return object means you can check http://www.cncmods.net/files/dbus/net_interface.xml . There are dbus bindings for various programming languages so you shouldn't have problems getting the same results programmaticaly ;)

Edit: oh, wait, did you mean "I have no idea" or really "I have now idea"? I missed the "w" in "now" somehow, excuse me... :o

ade 2013-12-10 21:06

Re: [Announce] Cellnet-Info for N900
 
Sorry, I meant "no idea".

I know this command, but it does not contain the used gsm band :(

misiak 2013-12-10 21:57

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1394683)
Sorry, I meant "no idea".

I know this command, but it does not contain the used gsm band :(

So I got you correct despite the typo ;) You are right, I don't see it either in any return object of any method listed there... I also tried readelf'ing /usr/lib/csd/plugins/libcsd-net.so, but no symbol with word "band" in it, nor in linked libraries detected by ldd... Will try to search deeper (something like leaving "find / -iname *.so -exec readelf -Ws {} \; | grep band" for the night - but the command may be wrong or have typos as I just semi-randomly typed it now).

edit: in fact,
Code:

find / -name "*.so*" -print
might be better to list all shared objects in the system, and then "readelf" them like there is no tomorrow in search of "band" string.

Estel 2013-12-11 03:53

Re: [Announce] Cellnet-Info for N900
 
First of all, thanks a lot for your super-fast and detailed reply. As usual, an "ade's mark" of support quality should be stamped on your releases and visible from application manager :)

Quote:

Originally Posted by ade (Post 1394640)
I do not agree on the useless part. It's equal to the bar value in the status bar as far as I can tell. You must be blessed with a very good reception, as my signal strength varies between 100% and no signal at all from time to time. I doubt if "at+csq" results in a significant other pattern, but I will do some more measurements on that.

I indeed have very good reception here, but, apart from that, I think that "It's equal to the bar value in the status bar" is the core of problem. The thing is, that corporate software creators (be it for phone software, or cellular modem "partner software"), tend to assign completely non-standarized, authoritary and (sometimes) random values to what "100% signal strength" means.

That's why, for cellular-network tech oriented communities, those things are considered useless, and they reffer to -dBm *and* it's equivalent CSQ values, exclusively. While CSQ is not ideal (different modem manufacturers assign different things to CSQ=0, aka "no signal"), it is still more real-life useful, because CSQ=32 always mean *true* 100% signal strength (aka unreachable in real life), so any other CSQ - often calculated into %, by community software (as opposed to "vanilla" connection managers), gives some real insight on how close to that theoretical "ideal" our signal is.

Also, for our use case, having signal-bar from status menu cloned into % for cellnet-info, is (IMHO) less useful, than having "from zero to theoretical maximum", more precise measurement. Sure, dBm value next to % provides it, but why not have it calculated as % for "on first glance" experience, instead of doing math in our head every time ;)

BTW, if at+csq would give same results, it would mean that they really, *REALLY* screwed it. In no way anyone could ever get CSQ=32 (100%), and even CSQ=31 is rarely seen. Meanwhile, we're getting so-called 100% signal strength, as no rarity.

Cheers,
/Estel

// Edit
I really appreciate efforts to get band readings, attempted by more knowledgeable ones. I feel it little irritating, that ages-old nokia phones have super-precise readings from their hidden netmon program (do not confuse with N900's netmon in repos, which is just using the same name), including even *channel* (equal value to few MHZ that we're transmitting at), and we're struggling to get band :|

But, maybe that's one of limitations we must accept - then, even band response from modem would be nice to have.

ade 2013-12-14 14:02

Re: [Announce] Cellnet-Info for N900
 
Changelog 0.0.1-8:
Code:

* corrected dBm sign
* split provider and device info
* divert info added
* network time added
* Emergency numbers added

Remarks/questions:
  • My operator does not supply network time AFAIK, so can anyone who does receive this, check if a (valid) time is displayed using "Provider info"?
  • My primary device has 112/911/999 as emergency numbers, while my backup device shows 112/911/118/119/000/110/999/08. Can't relate it to regional or device settings. Any clue why they differ? Only difference is that my backup device has no SIM card, perhaps that's the cause.
  • opencellid.org is giving me issues retrieving cellinfo (failed to lookup cellid). Most likely opencellid.org related. This version should deal with that without errors, and you can always disable the internet lookup. So far, I have not heard they have changed their API.

@Estel:
These are some numbers I consequently measure in my area:
Code:

Signal strength    CSQ
100%            16,99
50%              10,99
40%              9,99
32%              8,99
20%              6,99
14%              6,99

So it does not give me real other insights in terms of signal strength fluctuations. Perhaps your CSQ values are a bit higher. As the collection of this info may also be harder inside cellnet-info, I won't include if for now, sorry.

Estel 2013-12-14 15:26

Re: [Announce] Cellnet-Info for N900
 
Quote:

Originally Posted by ade (Post 1396423)
[*]My operator does not supply network time AFAIK, so can anyone who does receive this, check if a (valid) time is displayed using "Provider info"?

My provider does supply network time, so I'll test it and report back as soon as autobuilder chew new version of cellnet-info (no update available here, yet).

Quote:

Originally Posted by ade (Post 1396423)
As the collection of this info may also be harder inside cellnet-info, I won't include if for now, sorry.

No problem, I was suggesting, only :) Just keep in mind that current values are really misleading, and your tests confirmed it - stating CSQ 17 as 100% signal is laughable, to say at least (of course it's not your fault, bullet at Nokia[?]).

In case you would like to fix it *without* query'ing CSQ from modem (which gives a little strange replies, anyway), I've prepared a simple algo to calculate CSQ from dBm (which you have access to, already). It is cellular variant, the most commonly used one (at least in cellular network communities):

Code:

CSQ 0 = 113dBm
CSQ 1 = 111dBm
....
CSQ X = 113-(X*2) [dBm]

Then, remaining thing to get % values is to assign percents to corresponding values (CSQ 32 = 100%, CSQ 16 = 50%, CSQ 0 = 0%, etc) and in-between things.

Thinking about that, you may skip calculating CSQ entirely - just check lowest and highest dBm values from algo above, and assign percents, from best to worst.

/Estel


All times are GMT. The time now is 01:08.

vBulletin® Version 3.8.8