How does EvDo work? (Possibly PRL related)

[Ascertion]

This image shows my local area, and the two bordering Sprint towers. My question is which of the two are actually the tower I'm connected to.

 

As of lately, I'm able to pull 2.4mbps connections over EvDo at my house, on usually a -80 dBm. But when I go to my grandmothers, signal is roughly the same but data speeds are around 550kbps. This leads me to the think that the northern tower is slow, while the western tower is fast. And when I go to my grandmother's, it connects me to the northern tower. I just don't understand how moving about a football field from my house to hers results in such different data speeds.

 

 

EDIT: Also, I've went to the northern tower (stood 20 ft from it) and data speeds peak out at like 400kbps. But I can't tell which site I'm connected to.

[NiteSnow]

This image shows my local area, and the two bordering Sprint towers. My question is which of the two are actually the tower I'm connected to.

 

As of lately, I'm able to pull 2.4mbps connections over EvDo at my house, on usually a -80 dBm. But when I go to my grandmothers, signal is roughly the same but data speeds are around 550kbps. This leads me to the think that the northern tower is slow, while the western tower is fast. And when I go to my grandmother's, it connects me to the northern tower. I just don't understand how moving about a football field from my house to hers results in such different data speeds.

 

 

You're likely connected to the same tower, but just on a different sector with more load. You would have to use something like Netmonitor to confirm which towers you're connecting to though.

[Ascertion]

You're likely connected to the same tower, but just on a different sector with more load. You would have to use something like Netmonitor to confirm which towers you're connecting to though.

This is possible? I thought the sector would share the same backhaul. Am I incorrect with this assumption? Apps like NetMonitor and Opensignalmaps tend to give me the wrong coordinates no matter where I'm at.

[WiWavelength]

This is possible? I thought the sector would share the same backhaul. Am I incorrect with this assumption?

 

Backhaul is a later bottleneck. The RF airlink itself is the initial bottleneck. And one sector most certainly can be more loaded than an adjacent sector on the same site.

 

Apps like NetMonitor and Opensignalmaps tend to give me the wrong coordinates no matter where I'm at.

 

You have the Galaxy S3 and access to excellent engineering screens. That should be your first reference. Third party apps are poor simulacra, if you ask me.

 

Check your EV-DO PN offsets in both locations, and report back.

 

AJ

[Ascertion]

Backhaul is a later bottleneck. The RF airlink itself is the initial bottleneck. And one sector most certainly can be more loaded than an adjacent sector on the same site.

 

 

 

You have the Galaxy S3 and access to excellent engineering screens. That should be your first reference. Third party apps are poor simulacra, if you ask me.

 

Check your EV-DO PN offsets in both locations, and report back.

 

AJ

Will do. Expect a post within the next 24 hours with screencaps.

[NiteSnow]

This is possible? I thought the sector would share the same backhaul. Am I incorrect with this assumption? Apps like NetMonitor and Opensignalmaps tend to give me the wrong coordinates no matter where I'm at.

 

It's odd that NetMonitor would give you the wrong coordinates, it pulls coordinates directly from the base station. Sprint's network usually provides extremely accurate coordinates for their towers. I've tested on Verizon in my area and they don't even broadcast coordinates. I guess it all depends on what market you're in.

 

I would have replied something not to dissimilar to what AJ has.

 

Spectrum usage, spectrum deployment and carrier load can, and likely will be completely different on each sector of the cell.

[Ascertion]

My house

 

Granny's house

 

What do you guys think seeing these screencaps?

[WiWavelength]

Sector is the same at both locations, but signal to noise is four times better at your house. That affects data rate.

 

AJ

[Ascertion]

Sector is the same at both locations, but signal to noise is four times better at your house. That affects data rate.

 

AJ

Which one tells you the noise levels?

[digiblur]

 

It's odd that NetMonitor would give you the wrong coordinates, it pulls coordinates directly from the base station. Sprint's network usually provides extremely accurate coordinates for their towers. I've tested on Verizon in my area and they don't even broadcast coordinates. I guess it all depends on what market you're in.

 

I would have replied something not to dissimilar to what AJ has.

 

Spectrum usage, spectrum deployment and carrier load can, and likely will be completely different on each sector of the cell.

 

Now that's funny. Net monitor has never given me the correct coordinates.

 

Sent from a little old Note 2

 

 

[WiWavelength]

Which one tells you the noise levels?

 

I am out and about so will have to explain more fully later, but SINR is "signal to interference + noise ratio." A 6 dB difference equals a 4x ratio.

 

AJ

[Ascertion]

Now that's funny. Net monitor has never given me the correct coordinates.

 

Sent from a little old Note 2

It's quite funny, before I knew that it gave me the wrong coordinates, I went to the points it said the site was at and saw nothing. Drove around for about 30 minutes looking for a tower, even rooftop panels and nothing. Waste of time lol. I finally figured out it gives incorrect coordinates shortly after joining this site.

 

EDIT: AJ, when you explain a bit more, can you also include whether NV will improve the noise/signal ratio?

 

Thanks

[NiteSnow]

Now that's funny. Net monitor has never given me the correct coordinates.

 

Sent from a little old Note 2

 

That's odd... Was LA a affiliate market previously?

 

Here's a screenshot of NetMonitor from my area.

 

 

 

What color is the satellite dish in the NID/BID list? If it's blue it's getting the location of the tower directly from the base station, if it's yellow it's getting the tower location from the Gears Geolocation API.

[boomerbubba]

It's odd that NetMonitor would give you the wrong coordinates, it pulls coordinates directly from the base station. Sprint's network usually provides extremely accurate coordinates for their towers. I've tested on Verizon in my area and they don't even broadcast coordinates. I guess it all depends on what market you're in.

 

Netmonitor (and several other utilities including CDMA Field Test using the Andoid telephony API) do report the coordinates broadcast by the base stations. But those are not always the actual coordinates of the base stations themselves. See this comment in the FAQ thread, and this thread in the Sponsor Forum: CDMA towers that squawk the wrong coordinates

[digiblur]

And to speak on the PRL thing, nope. The PRL has one record in there for EVDO for your area. All the record has is the nationwide Sprint Evdo subnet with no sectors defined. The record is actually the same all across the US the only difference is the various acquisition records are different based on the licensed spectrum in the area. And actually the subnet converts into Sprint's home SID of 4139 in KC.

 

The PRL means absolutely nothing for EVDO once your phone finds the tower. All it does is help it acquire it. As AJ previously described about airplane mode may get better speeds article. The tower takes over on the channels. Case in point there are 2 EVDO channels in my area that are not in the PRL but yet the phone uses them all the time.

 

The PRL does not specify which individual cell sites your phone uses. It specifies which cell systems your phone attempts to acquire.

 

Sent from a little old Note 2

 

 

[WiWavelength]

EDIT: AJ, when you explain a bit more, can you also include whether NV will improve the noise/signal ratio?

 

Your SINR at home is 8 dB, which is very good for EV-DO. Over the river and through the wood to grandmother's house, your SINR is only 2 dB, which is just okay. I have to wonder -- based on your map -- if grandmother's house is very near the boundary between the N and SE facing sectors of the site. If so, interference from the N sector will degrade signal from the SE sector, which is almost assuredly the serving sector in question.

 

As for Network Vision, it may improve signal strength. But such will also increase interference between sectors/sites, so any SINR improvement will likely be minimal.

 

AJ

[Ascertion]

 

Your SINR at home is 8 dB, which is very good for EV-DO. Over the river and through the wood to grandmother's house, your SINR is only 2 dB, which is just okay. I have to wonder -- based on your map -- if grandmother's house is very near the boundary between the N and SE facing sectors of the site. If so, interference from the N sector will degrade signal from the SE sector, which is almost assuredly the serving sector in question.

 

As for Network Vision, it may improve signal strength. But such will also increase interference between sectors/sites, so any SINR improvement will likely be minimal.

 

AJ

I would've never thought a sprint sector would be interfering with a sprint sector! Lol

 

Sent from my SPH-L710 using Tapatalk 2

 

 

[WiWavelength]

I would've never thought a sprint sector would be interfering with a sprint sector! Lol

 

Be thankful that sectors do "interfere" with one another. Though such is not entirely desirable, it is by design. If sectors did not overlap at all, then every cell would contain two or three several degree slices of no service.

 

AJ

[digiblur]

Be thankful that sectors do "interfere" with one another. Though such is not entirely desirable, it is by design. If sectors did not overlap at all, then every cell would contain two or three several degree slices of no service.

 

AJ

 

While we are on the talk about sectors... I've always found it interesting you'll see a tower with a config of 3-90 degree sectors. Then you'll see one with 2-90's and then 1-65 degree. None of them add up to 360 Guess they count on that bleed over.

[airb330]

Very interesting topic. I'm trying to determine some locations with Verizon....it seems they somehow mask any information that is relevant? The BID seems to be relatively useless and netmonitor, etc. don't even try to estimate where the cell site is located as seen in post #13.

 

I am trying to figure out why my one work location can get a -74 to -84db 1X signal and yet terrible data signals. Calls also, often, go straight to voicemail. AFAIK all of the local Verizon towers have LTE. However I can't seem to exactly know 'for sure' what site I am using. Any insight? Verizon, of course, claims everything is A-OK. Well my co-workers and myself are getting pretty irritated as we are all stuck on Verizon for work cell phones (State forces us on VZ).

 

SID: 08 (Delaware & Maybe MD)

NID: 08 (Same)

BID: I've seen 4705, 4706, 679 & mainly 4708. According to http://howardforums....d.php?t=1578315 these are useless anyways.

PN: I remember PN offsets from the old flip phone days but I cannot seem to determine them on my SIII.

 

http://opensignal.com/blog/2012/08/10/how-to-tell-in-which-direction-a-cell-tower-lies/ is an extremely interesting post but it doesn't seem to add up for my area. Not sure why that is exactly but my base ID's are certainly not in 256 increments.