Error in custom LTE discovery apps?

[orangeblue]

I noticed that LTE Discovery and SignalCheck seem to have a problem identifying B41 deployments that run on 8T8R's. Both show B25 even if you're connected to B41.

 

 

Is there a fix in the works?

[caspar347]

I noticed that LTE Discovery and SignalCheck seem to have a problem identifying B41 deployments that run on 8T8R's. Both show B25 even if you're connected to B41.

 

 

Is there a fix in the works?

There are live 88rs? Did I miss something?

[orangeblue]

There are live 88rs? Did I miss something?

 

Yes, like Las Vegas which has is one of the markets ran old Motorola Wimax equipment. All the new B41 deployment is on the new 8T8R radio.

[mikejeep]

Next time you believe you are seeing the error in SignalCheck, please send a screenshot to support@bluelinepc.com and also send a diagnostic report within the app by going to About > Send Diagnostics; please include your username and a brief description so I know what it's for.

 

Are you sure it's on Band 41 (i.e. confirmed thorough engineering screen)? SignalCheck uses the PLMN to identify B41, it should be 311490 or 311870. Band 25 and 26 use PLMN 310120.

 

-Mike

[orangeblue]

Next time you believe you are seeing the error in SignalCheck, please send a screenshot to support@bluelinepc.com and also send a diagnostic report within the app by going to About > Send Diagnostics; please include your username and a brief description so I know what it's for.

 

Are you sure it's on Band 41 (i.e. confirmed thorough engineering screen)? SignalCheck uses the PLMN to identify B41, it should be 311490 or 311870. Band 25 and 26 use PLMN 310120.

 

-Mike

 

Mike, I am very sure I am pulling data off a 8T8R running B41 equipment tower. I am until tomorrow in Las Vegas and Sprint has been very busy deploying 8T8R radios. I just found a couple of new towers just came online this weekend.

 

I have to disappoint you, but Sprint is using 310 120 for B41 in Las Vegas too. There is no Clearwire ID and everything runs on the same ID. I think I had posted a screenshot a couple of weeks back of a test tower but it seems as all their B41 towers run in Las Vegas on 310 120 which makes sense.  

[lilotimz]

Everything broadcasting from sprint equipment is broadcasting the same plmnid.

 

Different plmnid broadcasts are possible but not implemented.

 

Sent from my Nexus 5 using Tapatalk

[Trip]

I saw this message before I left for work this morning and here's what I see at three places on my way into work, and finally, at work.

 

 

 

 

 

Not sure if this is helpful.

 

- Trip

[danialgoodwin]

Yes, Trip, information directly from the engineering screen is very useful, thank you. One interesting note is how the PLMN ID in the engineering screen cuts off the trailing zero, I've only seen leading zeros cut off before this.

 

There are about five different ways via public Android API methods to get the PLMN ID, and at least three different ways of getting all of the other signal info. And, at certain times/conditions, each of the different methods may be more accurate than the others. In the next version of LTE Discovery, the debug emails will include all of the different values from the different methods so that we can compare them all across devices (feel free to also cc Mike, if he doesn't already have it in his diagnostics  ). By combining all the different methods in different ways, internal testing has already found this to be slightly more accurate than just using a single method throughout.

 

 

 - Danial -
Developer of LTE Discovery, along with Spence (SpenceSouth)

Edited September 23, 2014 by danialgoodwin

[Trip]

Danial,

 

I also noticed that, at least here in the DC area, the TAC numbers are in a different range for Band 41 versus Band 25/26.  I attempted to attach my log from SignalCheck Pro but CSV is not a permitted file type here.  But even in the pictures you can see the Band 26 one has a 4-digit TAC as opposed to the three Band 41 ones with the 5-digit TAC.

 

- Trip

[Trip]

I had been using ##33284# to get that engineering data, but I just found *#0011# and took a new set here at work.

 

 

On this one, it looks like the TAC gets cut off--two characters worth.

 

- Trip

[danialgoodwin]

Danial,

 

I also noticed that, at least here in the DC area, the TAC numbers are in a different range for Band 41 versus Band 25/26.  I attempted to attach my log from SignalCheck Pro but CSV is not a permitted file type here.  But even in the pictures you can see the Band 26 one has a 4-digit TAC as opposed to the three Band 41 ones with the 5-digit TAC.

 

- Trip

Great find!

 

CSV files are really just text files and can usually be opened by any text editor. To attach it, you may be able to just change the extension to ".txt" and the file will still easily be human-readable.

[mikejeep]

Everything broadcasting from sprint equipment is broadcasting the same plmnid.

 

Different plmnid broadcasts are possible but not implemented.

 

You mean going forward, or now? So are you saying that means the 311490/311890 broadcasts are from Clearwire equipment?

 

 

Yes, Trip, information directly from the engineering screen is very useful, thank you. One interesting note is how the PLMN ID in the engineering screen cuts off the trailing zero, I've only seen leading zeros cut off before this.

 

There are about five different ways via public Android API methods to get the PLMN ID, and at least three different ways of getting all of the other signal info. And, at certain times/conditions, each of the different methods may be more accurate than the others. In the next version of LTE Discovery, the debug emails will include all of the different values from the different methods so that we can compare them all across devices (feel free to also cc Mike, if he doesn't already have it in his diagnostics  ). By combining all the different methods in different ways, internal testing has already found this to be slightly more accurate than just using a single method throughout.

 

Yeah, the Samsung enginerring screens have some very interesting bugs like adding/dropping digits. I think it's funny that apps like ours are more accurate! And I also work with all of the different methods to get the PLMN.. and as you've probably noticed, every device and/or market seems to use them differently. And I get them all in my diagnostic reports too.. not that it seems to help!

 

-Mike

[lilotimz]

You mean going forward, or now? So are you saying that means the 311490/311890 broadcasts are from Clearwire equipment?

 

 

It has always been.

[danialgoodwin]

Yeah, the Samsung enginerring screens have some very interesting bugs like adding/dropping digits. I think it's funny that apps like ours are more accurate! And I also work with all of the different methods to get the PLMN.. and as you've probably noticed, every device and/or market seems to use them differently. And I get them all in my diagnostic reports too.. not that it seems to help!

 

-Mike

 

Yeah.. and here's the latest weirdest bug we've found and fixed. TelephonyManager.getAllCellInfo() is supposed to be available for all Android API 17+, but a Samsung device with API 18 didn't have that method, nor most of the usual methods that are supposed to be in TelephonyManager (checked via Reflection).

[mikejeep]

It has always been.

I didn't realize that. From what I interpreted from reading around here, and what I was told by Sprint engineers, I thought they were deploying on those IDs. Is there any other Sprint equipment broadcasting on B41, or is this another way to confirm an 8T8R site?

 

Yeah.. and here's the latest weirdest bug we've found and fixed. TelephonyManager.getAllCellInfo() is supposed to be available for all Android API 17+, but a Samsung device with API 18 didn't have that method, nor most of the usual methods that are supposed to be in TelephonyManager (checked via Reflection).

Even if the method is present, it doesn't mean the device is populating it.. considering the popularity if their devices, Samsung is way behind on radio goodies for us.

 

-Mike

[lilotimz]

I didn't realize that. From what I interpreted from reading around here, and what I was told by Sprint engineers, I thought they were deploying on those IDs. Is there any other Sprint equipment broadcasting on B41, or is this another way to confirm an 8T8R site?

 

 

Even if the method is present, it doesn't mean the device is populating it.. considering the popularity if their devices, Samsung is way behind on radio goodies for us.

 

-Mike

 

Sprint 2.5 is 310-120. Clear sites run 311-490/870. 

 

The statement to you were true back in the day when they were deploying only on Clears equipment but is not true with the 2.5 equipment they're deploying now.

 

Clear sites will be completely torn down and replaced by Sprint builds starting next year.

[RCM]

Band 41 going live in my area has PLMN 310120. I'm finding that the GCI of Band 41 is 1450 (hex) higher than the Band 25 GCI after removing the 2 digit market code and the 2 digit sector code. If you're not familiar with the GCIs in your area,  you might not notice you're on B41 here solely relying on SCP. 

[mikejeep]

Band 41 going live in my area has PLMN 310120. I'm finding that the GCI of Band 41 is 1450 (hex) higher than the Band 25 GCI after removing the 2 digit market code and the 2 digit sector code. If you're not familiar with the GCIs in your area, you might not notice you're on B41 here solely relying on SCP.

It's coming to SignalCheck in an update very soon..

 

-Mike