1.9 GHz LTE overload question

[tyboxer12]

Does an overloaded lte 1900 band have any effects on speed?

[jamisonshaw125]

Does an overloaded lte 1900 band have any effects on speed?

An overloaded band will always affect speed.

[tyboxer12]

I thought the higher the band the more capacity. Saying 800 has less capacity than 1900.

[Dkoellerwx]

I thought the higher the band the more capacity. Saying 800 has less capacity than 1900.

 

That is not the case.

 

It's the amount of spectrum used that affects capacity.

 

A 5x5 800MHz carrier has the same capacity as a 5x5 1900MHz carrier. The lower spectrum is usable slightly farther away from the site, and is better in-building, that's the difference. 

[S4GRU]

I thought the higher the band the more capacity. Saying 800 has less capacity than 1900.

People often mischaracterize why higher frequencies have more capacity. They don't have any more or less capacity per single megahertz.

 

The reason why people say that higher frequencies have more capacity, because it takes more sites to cover an area. More sites equals more capapcity.

 

Robert via Samsung Note 8.0 using Tapatalk Pro

[joshnys8913]

That is not the case.

 

It's the amount of spectrum used that affects capacity.

 

A 5x5 800MHz carrier has the same capacity as a 5x5 1900MHz carrier. The lower spectrum is usable slightly farther away from the site, and is better in-building, that's the difference. 

 

 

People often mischaracterize why higher frequencies have more capacity. They don't have any more or less capacity per single megahertz.

 

The reason why people say that higher frequencies have more capacity, because it takes more sites to cover an area. More sites equals more capapcity.

 

Robert via Samsung Note 8.0 using Tapatalk Pro

Awesome explanation guys! 

[tyboxer12]

Thanks for the explanation because I wasn't sure if it was bandwidth that had capacity & speed or if it was higher MHz that made things faster.

[WiWavelength]

People often mischaracterize why higher frequencies have more capacity. They don't have any more or less capacity per single megahertz.

 

Nuh uh, 800 mega has only 800 hertzes.  1900 mega has 1900 hertzes.  That's a lot more hertzes.  And the more hertzes, the better speed.

 

 

AJ

[miguell2]

Nuh uh, 800 mega has only 800 hertzes.  1900 mega has 1900 hertzes.  That's a lot more hertzes.  And the more hertzes, the better speed.

 

 

AJ

I like my bands to have more hertzes than less. Just like my sandwiches, the more hertzes the better I always say 

[caspar347]

Thanks for the explanation because I wasn't sure if it was bandwidth that had capacity & speed or if it was higher MHz that made things faster.

If you're interested in a longer answer, here's my understanding of it:

 

Speed is affected primarily by the size of the channel, the quality of the signal, and the number of people connected. The latter two are pretty simple, the more people on a tower the lower the speed will be and the worse your signal is the worse the speeds will be.

 

Channel size is the tricky one to understand. Spectrum is sold in blocks by the FCC in various sizes, like 5x5 and 10x10. A 5x5 is two 5 Mhz chunks of spectrum that are sold together. Sprint's current LTE pairs are as follows:

 

• 800 Mhz: a single 5x5 or a single 3x3 (but we haven't seen the 3x3 yet)
• 1900 Mhz: a single 5x5 in most areas, two 5x5s in a few areas (like Chicago)
• 2600 Mhz: this spectrum works a bit differently, as it doesn't work in pairs. Sprint has, in most areas, at least 5 20 Mhz chunks, in some places they have more, in some places they have less.

 

The bigger the channel size, the higher the speed. A 5x5 goes up to 37.5Mbps, a 3x3 goes up to 22.5Mbps, and a single 20 Mhz chunk of 2600 will do about 85Mbps max.

 

This means that, theoretically, 1900 and 800 should have equal speeds in most places because both are 5x5s. But in real world scenarios, physics do not allow for it. As you probably know, 800 stretches further than 1900. This means that 800 is able to reach more people than 1900. It also can not be on every tower, as interference would cause issues. This means that generally, more people will be on an 800 channel than on a 1900 channel. So 800 will usually be slower. And 1900 will usually be faster because, going back to Robert's answer, there are more towers per square mile broadcasting 1900 than 800.

 

To use another carrier for an example, Verizon has a 10x10 on 700 Mhz. Because the channel size is 10x10, the max speed is 75Mbps, much faster than Sprint's 37.5Mbps 1900. But, 700 covers more people per tower than 1900, which means that 700 can get overloaded very quickly if too many people are connected. 1900 partially avoids this issue because less people are connected to any given tower.

 

So, in conclusion, the speeds are affected by channel size and how many people are on a channel or tower, not by the frequency number. However, the frequency indirectly affects speed, since higher frequencies require more towers to cover an area.

 

I could be wrong about some of this stuff, and I hope that some of our more knowledgeable members will correct me if this is the case. But I hope this helps.

[tyboxer12]

Thanks! All of these answers are very good explanations.

[tyboxer12]

Does anyone know about a B25 bandwidth expansion from 5x5?

[caspar347]

Does anyone know about a B25 bandwidth expansion from 5x5?

Currently unlikely, as some of the first few LTE devices only supported 5x5, not 10x10. However, Sprint is planning on activating a second 5x5 to run alongside the first one, so it'll be like the 10x10 but without the high peak speeds. It'll have the same capacity as the 10x10. Half of the devices in the area will be on the first 5x5 and the other half will be on the second.

[anthony.spina97]

I could be wrong about some of this stuff, and I hope that some of our more knowledgeable members will correct me if this is the case. But I hope this helps.

This is an excellent post! Nice job!

 

-Anthony

[WiWavelength]

Nuh uh, 800 mega has only 800 hertzes.  1900 mega has 1900 hertzes.  That's a lot more hertzes.  And the more hertzes, the better speed.

 

See, lots of hertzes.  And they got fast speed.  I proof my point.

 

 

AJ

[nowerlater]

See, lots of hertzes. And they got fast speed. I proof my point.

 

 

 

AJ

Proof? Interesting! How does one proof his point?

 

sent from my HTC One Max

[modplan]

 

This means that, theoretically, 1900 and 800 should have equal speeds in most places because both are 5x5s. But in real world scenarios, physics do not allow for it. As you probably know, 800 stretches further than 1900. This means that 800 is able to reach more people than 1900. It also can not be on every tower, as interference would cause issues. This means that generally, more people will be on an 800 channel than on a 1900 channel. So 800 will usually be slower. And 1900 will usually be faster because, going back to Robert's answer, there are more towers per square mile broadcasting 1900 than 800.

 

 

I was under the impression that b26 would go on every tower, and that downtilt and transmission power would be used to avoid interference with nearby towers, since the current towers are spaced for 1900mhz?

 

Basically they would strive to make 800mhz only fill the current 1900mhz footprint in metro areas, but giving us MUCH better building penetration on b26. And in rural areas where interference of nearby towers was unlikely, they would let b26 go as far as it could.

 

Is that wrong?

[caspar347]

I was under the impression that b26 would go on every tower, and that downtilt and transmission power would be used to avoid interference with nearby towers, since the current towers are spaced for 1900mhz?

 

Basically they would strive to make 800mhz only fill the current 1900mhz footprint in metro areas, but giving us MUCH better building penetration on b26. And in rural areas where interference of nearby towers was unlikely, they would let b26 go as far as it could.

 

Is that wrong?

You are correct. I'm referring to extremely tower-dense urban areas where congestion is a huge problem. If a site is only a block away from another, only one really needs B26 and they may interfere if put on both.

[S4GRU]

Nuh uh, 800 mega has only 800 hertzes.  1900 mega has 1900 hertzes.  That's a lot more hertzes.  And the more hertzes, the better speed.

 

 

AJ

 

It hurtzes my head to read that.  

 

Robert

[WiWavelength]

Proof? Interesting! How does one proof his point?

 

Measure the alcohol by volume percentage of his point, then multiply by two.

 

AJ

[LoneSnark]

The statements in this thread don't feel accurate to me. Sending data over radio waves involves putting a sequence of symbols on the available channel space and transmitting them long enough for the receiver to determine what symbol is being sent. How long a single symbol needs to be sent is dependent upon the modulation frequency, as it should take time, measured in cycles, for the receiver to identify when the frequency and phase of a modulated signal has been changed to. Therefore, doubling the modulation frequency should allow you to shove twice as many symbols on the channel per second, and therefore twice the data per second. What is special about cellular telephony that this is not the case? 

[centermedic]

I like my bands to have more hertzes than less. Just like my sandwiches, the more hertzes the better I always say 

Wouldn't that be wiches? I try to stay away from wiches. Especially wicked ones from the west.

[WiWavelength]

The statements in this thread don't feel accurate to me. Sending data over radio waves involves putting a sequence of symbols on the available channel space and transmitting them long enough for the receiver to determine what symbol is being sent. How long a single symbol needs to be sent is dependent upon the modulation frequency, as it should take time, measured in cycles, for the receiver to identify when the frequency and phase of a modulated signal has been changed to. Therefore, doubling the modulation frequency should allow you to shove twice as many symbols on the channel per second, and therefore twice the data per second. What is special about cellular telephony that this is not the case?

 

No, the info in this thread is accurate. PSK or QAM symbol duration is not at all frequency dependent. For LTE, it is a fixed duration -- as that is required to maintain orthogonality among all subcarriers.

 

AJ

[Dfarley]

No, the info in this thread is accurate. PSK or QAM symbol duration is not at all frequency dependent. For LTE, it is a fixed duration -- as that is required to maintain orthogonality among all subcarriers.

 

AJ

You said orthogonality just to make me feel stupid, didn't you?

[WiWavelength]

You said orthogonality just to make me feel stupid, didn't you?

 

Trust me, you like orthogonality.  Though in different ways, it is crucial to both CDMA1X and LTE.

 

AJ