We have previously learned that a millimeter wave is one of the fundamental techniques that 5G can use to remarkably improve its performance over legacy technologies. And we are at a point where we can meaningfully learn what 5G millimeter wave is. So, that is exactly what we'll do in this module. We will first take a quick look at what different 5G frequency bands are possible. And then we will focus on one of those bands, which is millimeter wave. We will start by learning a little bit about what are some of the challenges that the millimeter wave operation might pose. But on top of that, what are some of the fundamental benefits that millimeter wave will offer, which will make its commercialization and deployment economically, as well as technologically viable. And then we will look at some of the deployment scenarios for a millimeter wave. First of all, we will look at some of the conceptual scenarios that are possible. And then we will end the discussion by looking at some of the real-life examples of millimeter wave deployment. Which will drive home the point that millimeter wave is not just a pipe dream, but it is indeed a commercially and practically viable method of deploying 5G networks. So, let's get started. This is a quick summary look at the three principles frequency bands in which 5G as it is currently designed can operate. First of all, we have the proverbial low bands that are under 1 Gigahertz, also informally known as Sub-1 GHz. Some of the example bands in which 5G could be deployed here are bands like 600 MHz, 700, or 850 MHz, something such. Now we have seen previously in module one that as the frequency of operations increases the path loss or transmission loss that the wireless signal encounters also increases. Meaning that at a lower frequency path loss that the wireless signal undergoes will also be limited. And that is a one crucial reason why such low bands tend to be popular, not just with cellular services, but with many other communication services as well. The flipside of the coin is that because those bands are popular, there are many, many players contending for a little sliver in that frequency spectrum. And as such, the bandwidth that a certain occupant might get, for example, a certain network operator is pretty limited. Typically, it is maybe 10 to 15 MHz or maybe 20 MHz at the maximum. And that brings us to broaden horizon a little, because as we know, it's okay to have such a low bandwidth. But for truly significant performance improvement, we are also going to have to look at some of the broader channels. That is what brings us to the proverbial mid bands, also known in 5G parlance as sub-6 GHz, because these are the frequency bands under 6 GHz but above 1 GHz. The most prominent band or the frequency range in this spectrum is about 3.5 or 3.6 GHz, which is a part of the wider channel 3.4 to 3.8 GHz spectrum. Because this frequency spectrum offers us a wider channel bandwidth, we can think of this spectrum to deploy some of the 5G specific use cases or service classes, such as eMBB and mission-critical services. Typically, if not always, the amount of channel bandwidth available to a given cellular operator in this frequency range can be as high as 100 MHz, which is significantly higher than what was available somehow. And the third frequency band in which 5G can operate in under its current design, it's called millimeter wave. And it is the range of frequencies above 24 GHz. Most notably the frequency ranges, I should say that are currently picked for 5G's deployment are around 28 GHz and around 39 GHz. And in fact those happen to be the most prominent frequencies of 5G deployment in North American markets. In other markets such as European, or Chinese, or East Asian markets, some of the substance frequencies are also popular around 3.5 or 3.6 GHz. So, these are the three principal frequency bands in which 5G can be deployed. Now keep in mind that the 5G technology at such doesn't change dramatically depending upon what bands it is deployed in, some of the underlying principles and techniques remain applicable regardless of the bands. Although, depending upon the band of interest, there maybe some optimizations you can make along with the inherent advantages of those bands. That can give us significant performance improvement in 5G deployed in one frequency band as opposed to another frequency band. Although at the end of it, it is still the same 5G technology deployed in different band options. Now, regardless of what specific band you are deploying 5G in, it also matters how you're using the spectrum for 5G or any other technology for that matter. And there are three principal ways of utilizing the available spectrum as far as 5G is concerned. And those are given over here at the bottom of the page. First is licensed spectrum, and it is called so because whoever wants to use that 5G spectrum has to have a specific license of usage, obtained from a local regulatory authorities, which in the US, as we discussed would be the FCC. So, because whoever has the license to operate in that band has exclusive access to that band, no one else can utilize that frequency band. Stands to reason that the performance that can be achieved in such bands tends to be remarkably good. So for example, if there is a certain cellular network operator that participated in a FCC auction and won the licensing rights to a certain frequency band, that cellular network operator would have the exclusive rights to use that band. It will be illegal for anyone else to transmit or receive in that band. And because that operator's customers are the only ones in utilizing that frequency spectrum, their performance would be better than otherwise. However, as it is an engineering problem, you can be sure that there are trade offs. In that, although performance is better and it's guaranteed in the licensed spectrum. The network operator also has to shell out a lot of money, millions of dollars, if not more, in order to purchase the exclusive usage rights to a certain frequency spectrum from the FCC. And ultimately, those costs get passed along to the consumers. And that is a chunk of what we pay our regular, a monthly cellphone bills for. At the other end of the rainbow, there is unlicensed spectrum, which is the exact opposite of licensed spectrum. In that this particular spectrum, it has been internationally agreed upon to be free for use for end consumers. You don't have to pay anybody any licensing or usage fees, if you want your device to operate in such an unlicensed spectrum. Think of your Wi-Fi router, or your bluetooth, or your garage door openers, your baby monitors, etcetera. There are plenty of wireless devices around us that we don't have to pay anybody usage or license fees for. And the fundamental reason behind that is because they operate in unlicensed spectrum, most notably around 2.4 GHz or around 5 GHz in frequency. And because it is unlicensed, it is free. So, you don't have to spend any money to buy or allocate spectrum. However, because the spectrum is free and many other people would be trying to use the same spectrum at the same time. Imagine a densely packed neighborhood, let say a tall building where in multiple homes on every floor are trying to utilize their own Wi-Fi network, which nonetheless operate in the same frequency band. Given that there could be multiple users trying to use the resources in the same unlicensed spectrum at the same time in nearby spaces. You know that the interference profile in unlicensed spectrum is usually very poor, poorer definitely than what it would be in a licensed spectrum. And that is why we cannot guarantee performance in unlicensed spectrum. And that as you may correctly guess, flies directly in the face of some of the initial promises 5G has made in terms of guarantee in latency, through ports, reliability, etc. Is there a middle ground between the two? Well, as with many other engineering problems there absolutely is. And that is the paradigm, newly devised these days, the paradigm of shared spectrum. In that the underlying spectrum is still licensed, somebody has to license that spectrum from the FCC, for example and pay the upfront fees. But after that, it is not necessary that only one competitor or only one player can use that spectrum. For example, multiple companies or multiple network operators can hash out an agreement between them, in order to carefully synchronize and coordinate between them. So that they can utilize the same spectrum without interfering with each other. For example, if there are two companies that are trying to buy a piece of the spectrum in a shared paradigm, let's say they work out an arrangement between them as an example, let's say one company will use top half of the spectrum and the other company will use the bottom half. Or for example, one company will use the entire spectrum in the morning, whereas the other company will get to use the whole spectrum in the afternoon. These are just simple examples, but more complex arrangements can always be chalked out. And that is the fundamental paradigm of shared spectrum. So, the basic disadvantage of licensed spectrum, its cost is cut by a significant margin when you share spectrum with multiple players. But at the same time, because it is still licensed spectrum at its heart, you don't have any disadvantages with respect to unlicensed spectrum. Because you have only a limited number of players to contend with, and its performance goes without saying will be somewhere between that of licensed and unlicensed spectrum. So in that sense, it offers a good trade off between cost and performance. And now that we have seen some of the fundamental ways of utilizing or allocating 5G spectrum along with what are the three principal bands in which 5G can be deployed in the short-term. Let's try to focus on a frequency band that has perhaps garnered the most attention for very good reasons, the band of millimeter base.
