boomerbubba

What I do is go to the ##DATA# (##3282#) menu from the dial pad, then select View Mode->LTE record->LTE available file. You can get the BSID of the LTE-enabled towers that you have connected to. Once you have those numbers, you can correlate them to the towers that you see on NetMonitor.

That functionality apparently is available on the EVO 4G LTE, but it does not exist on my Galaxy GS3 -- or, I believe, on other devices..

BSIDs are IDs for CDMA, not LTE. I don't think we know what the undocumented EVO screen means. I think it might just mean the CDMA sector IDs that were detected while the phone was concurrently connected to an LTE radio. But that does not necessarily mean that the CDMA tower and the LTE tower were the same place. We just don't know for sure.

What tools are people using to monitor your connections? Speedtest and netmonitor are two obvious ones but are there *good* other ones?

It depends on what you mean. Speed Test is commonly used for testing data speeds, but it does not tell you where the handset is connecting to. There are several apps that will show and record CDMA connections, and even coordinates with complex limitations. In conjunction with the S4GRU sponsor maps of where almost all the towers actually are, such apps can be useful. I have used Netmonitor, but stopped because I don't like its privacy risks. I prefer CDMA Field Test for that purpose. But there is no known app that will map an LTE connection directly. Many here use Sensorly to track LTE signals, but it maps where the user is, not where the towers are.

Just picked up a 4G connection Dempster and Waukegan Rd. 20 meg up, 8 down. Net monitor ids the location as Morton Grove, School St.

 

4G continues up Waukegan going north. Tower location Morton Grove, N Waukegan Rd.

Welcome to S4GRU. This is a recorded reply, as similar information has been posted here countless times.

Netmonitor does not show you where your LTE connection is coming from. It only shows CDMA sites, and even those might not be the actual locations. If you want to see where the actual Sprint towers are on a map, become an S4GRU sponsor. Even then, it is tricky to identify where your actual LTE signal is coming from in the field, but it helps with the analysis.

Ok Austin Folks,

 

I just got a small but valuable nugget of information from the Sprint Store in South West Austin, Located at Mo-Pac and 290.

 

The guy there told me there is only one tower in South Austin with LTE turned on at this point and its someplace on S. Lamar. So any signal hits in that area of town for a bout a 5 mile radius are the same tower source. (A Pretty Impressive Range!) EDIT: Looking at my other screen captures for the signal strength confirms the same tower. He also said there were 2 confirmed LTE towers on in the North Austin Round Rock area but did not know more precise locations. Once LTE goes officially live (no date given) There will be many more towers. That will have it and signal will not be an issue.

 

While talking with the guy my phone would go in and out of 4g inside the store.

I did get a screen shot while in the DEBUG menu to capture my signal strength. Which was WEAK.

I think that description is not far from the truth, and I believe the number of live LTE towers is still pretty small. The Sensorly tracks are thin and those tracks are flaky on the ground. But I will quibble a bit. Over in the Sponsor thread, where we can freely discuss tower details from the interactive S4GRU maps, we have empirically detected at least three live towers in North Austin and Round Rock, although only two of them have been pinpointed geographically. In addition, we have documented a few other towers photographically with NV/LTE equipment mounted that are not yet live.

boomerbubba - thanks for the detailed comments. Let me address them one by one:

 

1) I understand your frustration with this. We're working on a series of accuracy improvements as we speak that will utilize a lot more accumulated information to improve the accuracy. There are a lot of challenges with this data and its a work in progress but we're working towards a longer term goal and I think its worth it. We are looking at better ways of communicating this to users, one of which is a confidence rating which takes into account the amount of data we have on a given tower and a series of different parameters.

 

2) Can you clarify what is meant by "lat/lon coordinates are not current"? You mean that the location records simply relate to older locations rather than the current? If so, would like to look into this in detail.

 

For your other comments we are very interested in getting all radio information available. As we understand it there are some difficulties though and not all handsets provide access to LTE signal strength and LTE tower information. Are you getting different results? We've found methods available on some devices but its not consistent across handsets and our users are on a lot of different models.

 

Please keep the comments coming!

As for the lat/lon coordinates, my experience was limited to one day testing. When I looked at the logs, it did not appear to me that the lat/lon coordinates were being updated for the actual location of each signal record even though I had been moving around. Identical coordinates just seemed to repeat for many records in a row. Maybe this was a bug, not bad design.

Going forward, tf you do separate the conflated records for different type of data connections, there certainly is a possibility that not all fields can be populated for all records. And the raw values for dBm are calculated and reported differently for different connection types (RSRP vs RSSI), etc.) So any generalized reporting would have to accommodate those and distinguish them in the logs.

I understand your difficulties getting the LTE signal data. AFAIK, Google's Android telephony API still does not have a neat, standardized method to expose them. So I guess until that happens, app developers have to dig into the non-standard APIs offered by different OEMs. (I have seen some callback methods to the hidden APIs described as a workaround by some app developers, but I think they are discovered more than advertised by the OEMs.)

I thought you must at least be doing that kind of workaround for LTE signal strength on major flagship handsets like the Samsung GS3 and transforming that value into your generalized column for signal. Because there are values for signal in those logged LTE records. Are they really just coming from the concurrent CDMA connection? That is even worse. Wrong data is not better than null data when the content is unknown.

I also understand that there is no standard ID value and location per base station available in the Android API for a carrier's LTE radio. These values, which are not well understood, seems to be displayed differently within the hidden menus of different OEMs' implementations.

Founder of OpenSignalMaps here,

 

Not sure if you have seen our app at all, but we are big fans of your site. We'd be very interested to hear if you find our app useful for tracking the 4G roll out or if not - what could we do to make it better? We're open to any constructive criticism at all & are constantly working to refine both the app and the information on the website.

 

We're currently working on an overhaul of our android application which we are pretty excited about - should be out by the end of October.

 

Brendan

1) I suggest that you lose the "towers" imputed mapping function, which is wildly inaccurate disinformation. We have to spend a lot of time debunking it for newbies here who see this stuff on your site and believe this is where the towers actually are.

2) Fix the log your app produces locally (I think it is called BSLocationLog.csv) to include useful information. Right now, its lat/lon coordinates are not current, it conflates the CDMA base station IDs with LTE signals giving the false impression that the LTE signal and the concurrent CDMA connection are coming from the same place, and garbles the signal strength data into some muddled metric that does not relate to raw data collected on the Android.

If you provided a log file with accurate signal strength for LTE signals (in dBm RSRP), along with accurate and timely coordinates, that would be useful information. Including an ID that actually corresponds to the the LTE radio, such as the Serving Cell information available in the LTE Engineering screen of the Galaxy S 3, would be a huge plus.

As it is, I tried using OpenSignalMaps for logging, and uninstalled it.

I've done that on my EVO 4G LTE and it just displayed no bars, period.

Thats odd, my GS3 actually shows bars with it turned on.

Nothing odd, really, because there is no standard whatsoever. Handset OEMs, carriers and even custom ROM developers are free to display whatever "bars" they want, based on whatever metrics and intervals they want.

Bars = BS.

I installed this latest update several days ago, and this week I have been doing a lot of LTE field testing. I'm afraid that some vestige of the old problem remains after all the firmware fixes: Sometimes, after losing an LTE connection and then moving into an area with an LTE signal, the handset will not automatically reacquire 4G status. But toggling the data connection by one of several means might force a renewed 4G LTE connection.

I was fooling about with Google Fusion tables and exported my LTE speed tests to a fusion table if people wanted to look at the results they are here: https://www.google.c...l2&y=4&tmplt=-1

(hopefully people will be able to see the map - the red dots are each a speed test that I exported from the speed test app - click on the dot to see the details.)

How accurate are the coordinates when you zoom in and look, relative to your actual locations? I have found that Speed Test sometimes does not capture them very closely, which I think may be a function of how often the app refreshes the Android GPS fix.

Just a suggestion: Creating this thread in the public Network forum will cripple it, because that will omit comments by Sponsors who have knowledge of where the towers are and wish to discuss empirical speed tests in that context. That discussion of tower-related detail belongs in the Sponsor forums. Become a Sponsor and see.

anyone else having issues with sensorly? i logged a 50 mile journey travelling thru all of the already marked lte zones, and not a single lte spot was recorded on my phone. i stopped the car, and toggled airplane mode and bam, 4g. once i started driving it switched to 3g. i was able to do a speedtest in a 4g zone and got .26 down and .07 up. what's wrong with this picture??

Some of the new LTE signals may have been turned off or blocked after a day or so. I logged some places yesterday not previously recorded, but a couple hours later could not replicate earlier sightings. Remember, the Austin market is not officially live yet. Such behavior is consistent with what we have read about in other developing markets.

I was just in the google play store and ran across this app called a network signal speed booster. How does this work and is it really relevant?

There are several apps in the market fitting this general description, all of which grant the app permissions to harvest some personal information, including phone call connections for example, and possibly upload it to some unknown server somewhere. As for improving your signal, I suspect all they do is basically some placebo action that resets the connection. (You can do the same thing just by toggling airplane mode.)

I also don't believe the reviews, which can be phony, generated by online "marketing" companies.

I usually get 1x at my house from these towers on Vashon and I was amazed to see a 3g signal. I checked netmonitor to see if there was a new tower around but it was my usual and hopefully former 1x towers. Is this a sign that something is getting started? And if I'm getting 3g from these towers, when lte is up, do you think it will reach me, I live down a pretty big hill?

You cannot rely on Netmonitor to show you towers directly without a master map of all the towers to inform your analysis. (There are many threads around this site explaining why.) The best place to find such a master map, showing almost all Sprint towers, is to become a Sponsor here at S4GRU.

Fixed the i/I problem while typing mid-word

Samsung must be confident in its lawyers. Everybody knows that the reason the previous version disabled the lowercase letter "i" on the keyboard was that Apple had patented it.

FYI, if anyone wants to apply the L710VPLI3 update patch file directly on a stock GS3, here is a link to download it. (Hat tip: xda)

Also looks like it turns on the Connection Optimizer....hmmm....prompts EULA and reenables itself as soon as you uncheck....

Whenever you go to that screen, it does recheck box to enable the feature by default. But if you uncheck it and leave the screen, I think it stays disabled. This is easy to test at your home WiFi location, which presumably is remembered with your WiFi login credentials, with WiFi disabled on the Notifications bar. When I do that on my phone (and I have installed the LI3 update) my WiFi does not automatically turn on. At least not yet.

Would make sense, as I'm just north of there. Although NetMonitor is saying 2409 Mockingbird Dr

NID: 22 BID 2209 -78dBm

1) Netmonitor is telling you the CDMA 1x sector radio you are connected to, which is not necessarily the LTE site.

2) Even for mapping the CDMA towers, the coordinates reported by Netmonitor and similar apps may not be the actual tower site, but rather some distance offset around the tower. See this short primer on that anomaly.

ATX4G's hunches have been fruitful before. Since you are a sponsor, I suggest that you explore S4GRU's Sponsor-level map in that immediate vicinity and look for the precise location of any Sprint towers. (Commenting on those details should be confined to Sponsor-level forums here, not this public Network forum.)

Then explore around any such tower on the ground. Running Netmonitor or CDMA Field Test in conjunction with the S4GRU sponsor map can help you determine if this is the tower where your CDMA 1x connection really is. Then, if you still are getting an LTE signal, you can explore the vicinity with Sensorly to record the LTE connection.

Unfortunately, at this writing we really don't know what the "improved LTE connectivity" feature really does. And several hours after the Sprint announcement, I haven't seen a peep on any forum by anybody reporting that they have received the OTA yet. Maybe Sprint and Samsung are further tweaking the LTE threshold levels, or maybe this is something else altogether.

Not knowing what this new update is supposed to do must be frustrating for any hapless person who had just completed an extensive test of the LTE radio using the prior firmware.

I respectfully suggest that the comments about whether it is morally right or wrong to use LTE when Wifi is available have hijacked a useful technical thread that was created for solid, expository reasons. Perhaps a mod could break them off and transfer them to a separate thread.

I am curious, how does the wimax devices like evo 4g/3d show both wimax and 1x signal in the status bar?

It would be up to the ROM developer (the OEM in the case of a stock ROM). On my old Epic 4G, using the stock ROM with Touchwiz interface, the Wimax signal was depicted with a little dynamic 4G icon that had a few concentric "radio-looking" waves emanating outward. Not very precise.

Under the hood, the phone had a dedicated Wimax API completely separate from the standard Android telephony API, and the special Wimax API made the signal diagnostics available to app developers. Of course, mostly only Sprint phones had Wimax, so apps would have to be written with that in mind.

Sprint's Samsung Galaxy S III Getting A Small OTA (L710VPL13), Brings SWYPE, Better LTE Connectivity, And More -

 

Enhancements/Fixes

- Inclusion of SWYPE keyboard

- Improved LTE connectivity

- SMS Messaging improvements

- EAS sync improvements

 

"This update, which is build number L710VPL13 for those who like to keep up with that sort of thing, is rolling out in stages beginning now."

Interesting. They already did a big LTE optimization fix in the previous OTA (L710VPLG8), but some users still complain about even the new threshold setting. Maybe that tweaks that further to keep LTE connected with a weaker signal?

That's a good question. Since CM doesn't support dialer codes and you cannot access your Debug Menu/Engineering screens, its going to be very difficult to confirm whether the CM signal strength indicator is valid for LTE. It will likely require a second LTE device for confirmation.

 

Robert via CM9 Kindle Fire using Forum Runner

Don't the apps such as AnyCut and Quick Shortcut Maker work with the custom ROMs? These shortcut apps just bypass the dialer to launch all the same hidden utility apps.

Thanks Robert. There is no app out there that can read data strength and put it at the top of the task bar?

 

No. I am not aware of anything. The limitation is in the Android API. Unless this changes in Key Lime.

 

Robert

The difficulty for an app developer is the lack of a standardized API within Android itself for LTE. It is possible to get LTE signal strength by means of trickier calls to workarounds to collect the data in the first place to support different devices. But the workarounds are there. Otherwise, for example, Sensorly could not harvest LTE signal strength data for its purposes of uploading data to the Sensorly site. About the only LTE-related data that is available is the fact that an LTE connection exists, and its signal strength.

[EDIT: I may be giving Sensorly too much credit here. I have read elsewhere that maybe the app only logs a Yes/No value for whether an LTE connection (or any other type of connection) exists for a particular timestamp and location. That is, Sensorly ignores signal strength values and plots an arbitrary fuzzy area around these points. So I need further research to be sure of what Sensorly itself really does.]

[EDIT: A subsequent release of Sensorly does collect signal strength data.]

AFAIK, there is no related detail available for LTE covering other elements such as an identifier for the tower radio and its location, as there is for CDMA and GSM. So it is the logging function (in Netmonitor, CDMA Field Test, etc.) that is crippled for LTE.

If those bar on the phone are not data strength, in a way when you are near the tower and its 5 bar, then the LTE should be strong too in a way?

That would tend to be true -- but only if your LTE connection is coming from the same tower as your 1x connection. That may or may not be true in any specific case. It is less likely to be true today, when all towers handle 1x but only some handle LTE.

I've seen a network vision completion coverage map online Sprint‑4G‑LTE‑map.jpg and im wondering if this is with or without the 800mhz nextel spectrum.

I really don't think that low-resolution raster map would look any different at this far-out zoom level either way. But I believe that since it is supposed to represent "at completion of NV," that would include the 800 MHz coverage.

BTW, did anybody else read the blog post behind it? Pretty far off on its facts.