I have heard a lot about 5G and the roll out and how small cells will boom with 5G. Let’s look at the facts, if 5G is in the millimeter wave spectrum, then they will be very small and very line of site networks.
How is this going to work? Are the carriers really going to rely on 28GHz, (here in the USA), to cover much more that a building or a street? They are complaining about putting up small cells and CRAN because of the fiber and rental fees, and the slow payback! It’s catching on now and there is a better business plan, but will it be enough at 28GHz?
What I don’t’ get is that they are all doing amazing research and getting awesome speeds, but to me, someone who has done point to point (PTP) microwave, it is not a surprise. If you only have 2 or 3 users, of course you can get kick ass speeds in a high spectrum with lots of bandwidth that is not shared. At home, when I am the only one on my Wi-Fi hot spot, the cable modem is the bottleneck, not the Wi-Fi, however, when I am in an airport using the free Wi-Fi with 100 other people, I have no idea where the bottleneck is unless I can’t connect, then I am sure it is the Wi-Fi. It happens all the time.
So let’s look at the business model. Do you really see AT&T and Verizon building a network for 10 people? Let me put it to you this way, when you want to put a small cell in your building for 10 users, will AT&T or Verizon run out and give you one? I will tell you from experience, the answer is no, but they will let you buy one but it’s very small. Now, if I want to put something in for 100 users, will they do it. Again, the answer is no, and now because you want a bigger small cell or 2 of them, they won’t even talk to you because they are afraid of how it will affect the network. I get it, but now let’s think about how they will deploy 5G, will it really be for the mobile user? It doesn’t make sense to me.
Now, let’s look at fixed wireless, here is an awesome application. One that you need LOS, (line of site), but you have the spectrum and the ability to put a small cell near people’s homes. This is where I see it as a great asset. This would be great and finally would give some wireless competition to the Cable modem.
So for fixed wireless it looks like an awesome solution, but they have to make it work for the mobile device somehow.
The reason I am not so positive, because let’s look at path loss. At 200 meters there is a lot of free space loss here.
|Frequency||Distance (Meters)||Tx Output (dB)||Rx Gain (dB)||Free Space Loss (dB)|
The reason I added the chart is to show the high loss you have at the higher bands. Now, while the number looks like it is double the loss, it is actually more than that. For every 3dB you lose 1.2 of your power, literally, lose half of your power.
Take a look at this chart showing power in Watts compared to power in dBm, pay attention! It helps put it into perspective. I like to look at watts because it is easy for the field engineer to see loss in power.
|-30 dBm||0.0000010 W|
|-20 dBm||0.0000100 W|
|-10 dBm||0.0001000 W|
|0 dBm||0.0010000 W|
|1 dBm||0.0012589 W|
|2 dBm||0.0015849 W|
|3 dBm||0.0019953 W|
|6 dBm||0.0039811 W|
|9 dBm||0.0079433 W|
|10 dBm||0.0100000 W|
|20 dBm||0.1000000 W|
|30 dBm||1.0000000 W|
|40 dBm||10.0000000 W|
|50 dBm||100.0000000 W|
We need to understand what the application will be for 5G to be successful. Let’s face it, LTE would not have taken off without the demand for bandwidth. What caused demand for wireless bandwidth? You could say the laptop, but you know it’s really the smartphone, specifically the iPhone that changed the wireless world. Now we can’t live without it and even the president, (Obama) says it’s a necessity. Link is at the bottom so you know that I’m not making this up.
So how do we make money? The carriers will come up with something, but the mobility factors seems quite limited. The fixed wireless aspect makes a lot of sense to me, but if we can get 10Gbps to our mobile device, I am all for it, but who will pay for the backhaul? Latency is very important, so they have to bring the computing to the edge, NFV and SDN will be a huge part of 5G. The latency has to be low and then the bandwidth will not be as critical, right?
Who will win in 5G?
We have all heard about 5G, the next big thing, for the carriers anyway. Will it really be all what we think it can be? Super high bandwidth with awesome applications? Who will win the 5G race?
The way I see it Verizon Wireless and AT&T will win, or at least be there first. They invested heavy, they are working to put out real standards, and they really want to win. This is something that the other carriers seem to be playing with but not taking as seriously. Sure, they all say they are working on 5G and I am sure they have something in a lab with high bandwidth in the 28GHz range, or somewhere up there, to test these applications. That way the investors are happy along and the public is impressed. I mean we all want to see virtual reality happen very soon from a wireless device, right?
Verizon seems to be working hard on the technical standards and the testing of the network.
AT&T seems to be finding ways to use it and testing it as well.
While I see that Sprint and T-Mobile are looking at it and testing in labs, they probably are in no hurry to lay out a lot of money just yet. They may wait to see what standard will be adopted then take the money they saved and use it to build a network, or expand what they have. I hope they are planning for this future. In all honesty T-Mobile is doing a great job of laying the ground work for high speeds while Sprint, well, Sprint doesn’t seem to be doing much of anything for the future.
Will 5G be a Failure?
So will it succeed? The issue that I see right now is not the technology, the loading, the multiple access, but really the spectrum. How far can millimeter waves go? How many connections can 28GHz or whatever band they’re in, can it serve? How large of an area can it really serve? We all look to small cells or CRAN to perform here, but will there be a payback. Can the carriers make a play for fixed wireless or will they try to capitalize on dense networks? Can they make the dense network business model work? Does it have to be a carrier to build this or can it be a smaller business who could tie back into another company’s core? Will they need a mobile core or can it start to replace fiber? Remember, there were so many microwave hops connecting broadband when fiber was not so readily available.
I don’t think that the question of 5G is fair, to be honest, its more like, “Will Millimeter Wave be a failure?” Isn’t that the big question here? The elephant in the room, so to speak. Can mm-wave perform the way that today’s spectrum does? Can it go beyond PTP?
The real winner will be the fiber backhaul companies, they will offer maximum bandwidth, chances are it will be dark fiber dedicated to the carriers, one more expense the carriers to look at before deployment. This will be the big cost just like now. Sorry cable companies, but the cable modem may not be able to service such a huge capacity. Fi they really do virtual reality, they will need to get to the edge for low latency and have a lot of bandwidth and the devices will need to maintain connectivity. Or will they?
What is the plan?
Maybe this is where the devices will need to take a step forward again, like the iPhone. Then the networks pushed LTE out so that high-speed data could get the devices reliable. Then devices added more and more memory to improve it once again. This is where the 5G miracle may happen. The devices will really need to make another quantum leap. The way I see it is that the networks will dump data fast in large chunks for high bandwidth applications and the device will need to capture and process that data. The app will rest in the device and the device must take the data and break it down. We are doing this now with applications in our smart phones and tablets, but they need to be the edge, they need to process it all quickly and they need to be able to connect and accept data quickly and efficiently all while processing it for the application.
I really the think that the edge has to push farther out, so that the cloud will extend farther. I know they talk about fog computing, where the edge is at the end, this is what will really push the network. Hop after hop, so that the computing is done so close to the device that it may be in the device itself. Will there be awesome high-speed kick ass wireless devices that 10 other devices can connect to? I hope so. Will fixed wireless take over the world? I hope so. What are you hoping for, other than a bunch more work in deployment?
One thing that could hold it back is that the carriers do NOT want to replace their networks. They have LTE in as the foundation, they do not want to do another fork lift upgrade. They want to just keep making minor updates until it completely maxed out. They are counting on LTE to push them well past 2020. When we all see 5G released in 2020, it will be on the back of LTE, which technically is still 4G but by then it should be on serious steroids.
One other thing, the carriers do not want to give up their dominance. They intend to rule the wireless world well into 2030. So when the FCC has auctions, they will spend billions. Just look at how much spectrum AT&T is sitting on, you think they would roll it out soon.
Just a few ideas, but worth thinking about.
- http://users.ece.utexas.edu/~rheath/presentations/2015/mmWa5G business planveFor5GTWS2015Heath.pdf
So let me know what you think, email firstname.lastname@example.org when you think of something to say!
Be smart, be safe, and pay attention!
Do you know what to put in your SOW, the details needed to get paid for milestones or job completion? Would it hold up in claims court? Would you rather plan up front instead of fighting for it after the job is done?