Yes, you heard me, size matters! Why, because size and weight are what the tower companies will be looking at for the new massive MIMO antennas. Let’s call them active antennas to make things easy because massive MIMO will be a given for this article.
Why would size matter?
Let’s look at it this way, bigger antennas cost more money all the way around. Size has a direct correlation to costs. Let me break down what costs more with a bigger antenna.
- Larger antennas cost more to build. CapEx
- More radio heads cost more, CapEx.
- Larger antennas may need to have the tower structurally modified to hold the extra weight. CapEx on installation.
- Larger antennas may raise the monthly rate on a tower. OpEx on rent.
The point here is that there have to be a balance. The carriers know that payback has to balance out with the costs. That’s where we find balance, between the costs, CapEx and OpEx, and the payback, number of subscribers and improved performance. There has to be a set point.
These active antennas may not make sense to put everywhere. Do we really need to put them near a farm where there could be a total of 20 users at any given time? Probably not unless one of those users is a CEO or a president. Power and position has privilege.
We’ll look at what effects the size.
- Frequency matters. I’ll make this simple, the lower the frequency the larger the antenna. It’s that simple.
- TDD or FDD matter because with FDD you will have 2 sets of radio heads and TDD only has one. FDD will be bigger because 2 sets are larger than one.
- Size of massive MIMO, meaning the number of elements. If you have 32T by 32R, 32×32, you have 32 transmit and 32 receive elements. It doubles each time, 64×64 has 64 of each element and radio head, 128×128 has 128, and so on.
How has this changed from the traditional models?
When we had CDMA, FDD was all the rage. To have the dedicated spectrum for uplink and downlink made all the sense in the world. Then, with LTE, we thought it was nice to have dedicated spectrum each way, but the reality was that it became less of an issue and now with carrier aggregation and dynamic uplink and downlink balancing. Hey, Wi-Fi had it right all along. LTE is catching up to Wi-Fi’s lessons learned. Just like the MIMO technology, Wi-Fi had it first. LTE is putting that technology on steroids. Then 5G NR will amp it up even more. How cool is this?
I digress, sorry.
TDD or FDD?
If you think it doesn’t make a difference, it does. You see the carriers loved FDD in the CDMA world because they had efficiencies when the uplink and downlink. FDD allowed them to have dedicated uplink spectrum and downlink spectrum. This was a crucial factor for efficiency. Even with LTE is seemed to be a good thing when they had the spectrum broken apart. That is, until today, when the efficiency of uplink and downlink balancing was not possible when dedicated spectrum up or down may cause more problems than is solves. So now, Sprint’s 3.5GHz spectrum and the CBRS 3.5GHz spectrum looks quite sexy. It allows the carrier to control uplink and downlink dynamically free of any dedicated up/down spectrum barriers. Awesome!
For instance, LTE is more like Wi-Fi now. It can be more efficient when you can have the spectrum controlled by the carrier, not dedicated. I wondered if the carriers would think about trying to change their new spectrum. It seems like now, FDD having dedicate spectrum would create limitations. Wouldn’t it be nice to control what goes up and what comes down in LTE and especially in 5G. TDD allows that because it is transmitting, Tx, and receiving, Rx, on the same elements and in the same spectrum. How cool is that? Just like Wi-Fi, only it’s LTE, soon to be a 5G format. I would think 5G will be LTE on steroids.
Why does this matter in massive MIMO? Again, the FDD system will need dedicated antenna elements paired with radio heads for transmit and dedicated elements and radio heads for receive. Therefore, a 32×32 active antenna would have 32 transmit and 32 receive elements paired with a radio head port in the antenna which would effective look like, in my mind, 64 heads in one antenna.
A TDD system could have the receive and transmit together on one element. Therefore a 64×64 active antenna would have just 64 elements paired with radio heads.
The antenna that has half the elements should be half the size, smaller antennas with less weight make for a happier installation, lower costs, and more effective rollout.
Beck to cost, elements and tiny radio heads all cost money. The payback and gain by adding more active elements has to have balance somewhere. If 64×64 costs 5 times as much as 32×32 it may not be worth putting it in. If 128×128 costs 10 times as much, then when is the payback? There has to be a balance between antenna cost and system gain.
What about frequency?
How does this effect the antenna? Well, the antenna size is determined by the band. The lower the frequency the larger the antenna, or at least the elements. That’s a normal antenna. Now that we have massive MIMO, it makes more of a difference because the radio heads are behind each element in the antenna. This can be a factor in antenna size.
The lower bands, say 1.3Ghz and lower, are going to have larger antennas that require more size just due to the lower spectrum. That is if they want 3dB of gain or more. There are many factors with antenna design which I am not going to get into, but the lower the spectrum, the larger the antenna. Remember that the carriers want plenty of gain and need to have the efficiency to put the least number of antennas on a tower, say 3, as possible. If it is a mini macro on a pole or a small cell, then you may rely more on one or two antennas to cover what you need. Lower spectrum makes that more of a challenge.
While you think it may not matter, you’re not seeing the bigger picture. Larger antennas cost money and many carriers have spectrum in many bands. In fact, why do you think that T-Mobile wants the CBRS 3.5GHz spectrum to badly? They see the value in the short-range coverage. It’s high spectrum, smaller radios and antennas, and covers the smaller areas efficiently. The deal with Sprint fell through, now they need a contingency plan and the CBRS looks inviting.
How much is too much?
Here we have the real conundrum of massive MIMO. How much is too much? Do we know the payback of massive MIMO? It looks like we need it for true 5G to roll out with all the promise we expect of 5G. I mean it’s more than just the new format of 5GNR, it’s all the features that give us Ultra Reliable Low Latency, URLL, and extreme broadband.
There has to be a balance of where we put it, how we deploy, and so on. It makes sense to put it in urban area where the payback is immediate. Lots of users can justify the cost. If we are covering cows on an IOT system, then it doesn’t make sense, does it?
If the cost of a 64×64 is 1/3 the price of a 128×128, then it may make sense to go with the 64×64 for the payback. The number of radio heads will change the price of the unit along with the size and weight. We have to be financially responsible, don’t we?
Larger antennas cost more.
Then, there is the mounting issues. They will leave it up to the construction crews to install the equipment, but they won’t like putting monstrous active antenna on the towers if the tower companies raise the rent 10 times. They also have to consider the tower modification implications. There has to be a balance.
Now, for someone with a TDD system if they find the right model. If the model makes sense, then they could lighten the load on the tower. This may or may not make the tower companies happy, they want more rent but they don’t want to modify the towers if they don’t have to. Actually, they pass that cost onto the carrier, so maybe they don’t care.
For the FDD systems, they will have to install larger active antennas because the Tx and the Rx will be split. You need 2 active element arrays. This add size, cost, and complexity to the system. However, it will enhance performance of the system. You no longer need radio heads and coax jumpers since it is an active antenna.
But wait, that’s not the big picture!
The reality is, for mobility, we have to look at what we’re replacing. If the carriers are going to upgrade to massive MIMO in their existing spectrum and replace their existing equipment, then they have an advantage.
For instance, they will install one unit. The active antenna will have fiber running right to it, direct. So there is not longer all the crap on the backend, like the radio head, the coax jumpers, and a separate antenna. All of that equipment adds problems. Let me break it down, the radio heads used to have 1 to 3 fiber pairs running to them, that will change, now there will be many more. There is more data, more overhead, and more bandwidth needed. That is why all the fiber will be connectorized.
I know I threw a lot at you, but let’s look at everything and what it means.
- No more radio head, less room needed on the tower, the weight of the radio head is probably more than the radio heads in the active antenna. Less weight and one less point of failure.
- No more coax means less weight, no PIM testing, one less point of failure, no reflected power, easier troubleshooting, less time of installation. For those of you that don’t know, coax jumpers take a lot of time to make, weatherproof, tighten properly, and secure properly.
- Fiber connectors save a lot of time, in the old days tower crews had to put connectors on the fiber after they cleaned it and then test it thoroughly, all this takes a lot of time to install.
With everything in one unit, installation is quicker. Mounting should be easier. One unit to install, not many for each sector. However, now we have a huge point of failure, if the active antenna goes, we’re down hard for that sector.
One more thing, in theory, we should have electric downtilt with the massive MIMO antenna that will be controlled automatically by the system. So Azimuth is important but now we may not have to worry about the 3 degrees of downtilt like we used to.
Less time to install, easier to install, less equipment hanging on the tower. It’s a win-win all the way around. All this with increased performance. WOW!
Pros and Cons:
- Fiber to the antenna decreases installation complexity,
- Active antennas are integrated,
- Massive MIMO improves system performance for;
- Coverage through beamforming,
- Multi user, MU-MIMO, allows the beams to talk to multiple users simultaneously,
- Increased throughput to each user,
- Increased densification for power and throughput to multiple users,
- No more coax jumpers, PIM testing, weather proofing, and so on,
- Less weight overall due to less equipment on the tower,
- Increase system complexity,
- Increased cost for antenna,
- Could be a single point of failure, not sure about how the connection to the active antenna will work,
- More fiber jumpers up the tower,
- Probably increase power draw for the active antenna,
Things to think about?
- Cost of the array, does 32×32 serve your needs or can you go 64×64 or 128×128? Which delivers the best cost for the best price?
- If FDD, what size can you put ion the tower? Will it match the antenna size you have now?
- Are you ready to run more fiber up the tower or across the rooftop?
- Will the payback make sense?
How does the massive MIMO system payback the carrier?
- Increase throughput
- Much better densification, concentrating the power to each UE,
- Better throughput to each UE through beamforming and multiple users talking t the same time, remember that there are multiple radio heads behind each element,
- Less physical complexity on the tower,
- New options to carriers for deployment,
- In urban areas it could reduce the need for small cells in the macro’s coverage umbrella,
- CRAN Massive MIMO greatly improves localized densification,
- Spectral efficiency is greatly improved by the beamforming,
To learn more:
Let me know if this has helped you! Subscribe to this blog, at the top of the page or get me on Twitter @wade4wireless or email@example.com or go to www.wade4wireless.com or www.techfecta.com to reach me. I do have a podcast, search Wade4Wireless wherever you get your podcasts and subscribe. Reach out on LinkedIn, https://www.linkedin.com/in/wadesarver/ or Facebook, https://www.facebook.com/Wade4Wireless/ to stay in touch. I am very reachable!
I am building reports around these blogs for massive MIMO and 5G, soon to be released. They will be available in PDF and print, let me know if you’re interested in LinkedIn and send me a message so I can tell you where to get them. They should be released by April 1st.
For more information go to:
- I wrote an article about the differences between TDD and FDD, found here, https://wade4wireless.com/2017/01/16/an-overview-on-tdd-and-fdd-formats/
- If you want to learn more about Massive MIMO, https://wade4wireless.com/2017/11/27/what-is-massive-mimo/
- Learn more about massive MIMO beamforming at https://wade4wireless.com/2018/01/08/what-is-beamforming/ and https://wade4wireless.com/2018/01/29/about-massive-mimo-beamforming/
- The 5G eco system, https://wade4wireless.com/2018/01/22/the-5g-ecosystem/
- Basic antenna concepts, https://www.pulseelectronics.com/docs/library/Antenna%20Basic%20Concepts%2007%2012.pdf
- Antenna size and gain, bandwidth, efficiency, https://nvlpubs.nist.gov/nistpubs/jres/64D/jresv64Dn1p1_A1b.pdf
- Antenna basics, http://www.phys.hawaii.edu/~anita/new/papers/militaryHandbook/antennas.pdf
- Antenna design, http://bluflux.com/yes-the-600mhz-white-space-auction-will-impact-rf-design/
- Frequency bands, http://www.antenna-theory.com/basics/freqBands.php
- Why antenna size matters, http://aviatnetworks.com/media/files/Why_Antenna_Size_is_Important_Jul_11.pdf
Finally, one more thing:
I am winding down Wade4wireless because I am building up TechFecta. I have plans and I can’t do all of this at the same time. I want to build up a full-time business around this information and more. I will focus on tech, health, and philosophy. Those are the things that really fulfill me.
As you know, it’s exhausting to work full-time and do this on the side. While I really enjoy this, I have more that I want to do.
I would like to thank all of you for the support. I really appreciate it.
I will continue this for another few months, but I don’t know if I can maintain every week, it’s really a lot of work. Let me know what you think!
Be smart, be safe, and pay attention!
The foundations below do beautiful work, helping families in their time of need. Climbers often get seriously injured or die on the job. The foundations below support those families in their time of greatest need!