Now that we have seen how a general 5G network would look like, let's try to understand how would cellular network operators actually deploy the said 5G network. And when it comes to deployment, there are two options corresponding to the buzzwords that some of you may have heard of already. Those buzzwords being standalone or SA network or non-standalone or NSA network. But even if you haven't heard of those, that is okay. Here is where we will learn a little bit more about each of them. We'll start with the standalone option because that is somewhat easier to understand conceptually. What is a standalone 5G network? Well, at one end you have a 5G phone which connects to a 5G gNodeB, which itself is a part of a wider 5G RAN, and the 5G RAN connects to a 5G core network also known as 5G C. So end-to-end, it is a 5G system. And because that end-to-end 5G system is capable of standing on its own, so to speak, that is the reason why this particular option of 5G network deployment is called a standalone option. Now, this is where operators will want to be tomorrow, but this is not where they're at today. The scenario today is somewhat different. So how difficult is making the jump from today to tomorrow? Well, on paper, it sounds a little easy, but in practice, it is anything but. So let's try to take a schematic look at where the operators are today with respect to where they want to be tomorrow. So this is tomorrow's network. Whereas what they have today is a 4G core network, 4G LTE RAN, and all EUs. Now if an operator has to go from this scenario to this, then the operator, as we can tell, has to make a few key changes in that the operator would have to upgrade its core network from 4G to 5G. And even though it sounds easy on paper, it is quite difficult to do in the reality. It takes a few quarters worth of planning and a significant amount of budget for any operator to upgrade its core network from one generation to another. So this is not a turnkey solution, it is going to take some planning and some budget. Furthermore, upgrading the RAN from one technology to another is even more herculean a task. In that if an operator has let's say tens of thousands of base stations in a typical country like the United States, to go from 4G to 5G, operator has to upgrade the hardware and software on all of those base stations from 4G to 5G. And that is a non-trivial expense and effort in itself. But not only that, not only are the costs related to hardware and software, but keep in mind that in order for the RAN to operate in 5G technology, you have to have some 5G spectrum associated with that RAN. And you have to buy or lease that additional 5G spectrum, which operators currently don't have at least in this hypothetical scenario. So going from here to here is also an equally herculean task in terms of not just software and hardware, but in terms of spectrum as well. So although this is our idealistic utopian scenario that we would want all the network operators to have tomorrow, going from what they have today to this idealistic scenario of tomorrow is not going to be easy or overnight. And that necessitates a sort of a stepping stone, a middle ground between today's LT networks and the 5G networks of tomorrow. And that is where the NSA network step in. NSA, as you know, stands for non-standalone option. We will shortly see why it is called so. In a typical innocent network, operators can continue working with the existing 4G deployment. So operators who have 4G core network, which is called evolved packet core, will continue operating with that, and that 4G core network will connect with the nationwide deployment of 4G RAN. Now we are on the same page up to this point. But one question might pop up in some of your minds that, hey, one of the fundamental reasons, so we thought we had to move from 4G to 5G. Is that at least in certain areas, 4G may not be able to give you a sufficient amount of throughput network capacity and you will be exactly right. There are many areas such as commercial areas, downtown, football stadiums, concert arenas, etc, where the localized demand for throughput and capacity is so high that LTE by itself cannot meet the demand anymore. And that is where this NSA option comes into picture. The basic premise of NSA option is that the operators will continue working with their existing LTE network. But in some of those capacity challenged areas such as commercial areas, downtown, concert arenas or football stadiums, operators will selectively and additionally deploy a limited number of 5G base stations or gNodeBs. Unlike in a standalone option wherein you would have to have a nationwide deployment of gNodeBs. An NSA option allows you to have only a limited or surgical deployment of 5G gNodeBs in areas where you have severe capacity challenges which cannot be met by LTE. Now, once you have both 4G and 5G base stations in that area, something remarkable happens, you have a wireless coverage from two different technologies present in the same area. You have 4G signal coming in from the 4G base station and 5G signal coming in from the gNodeB. And that's where enters a special breed of modern phones which are called dual connectivity capable phones. They are called so because they are capable of connecting with two technologies at the same time, hence the name dual connectivity. A dual connectivity or DC capable phone can have a wireless channel with LTE base station. And at the same time it can have an independent parallel wireless channel with the 5G base station and those wireless channels will be LTE and 5G respectively. And because that phone has two wireless channels instead of one, the instantaneous throughput that the phone can leverage will be substantially higher than the throughput that an LTE or UE can obtain. And thus by increasing the instantaneous throughput or data rate or speed that is available to a given phone. The network operator will have pretty much alleviated or all together solved the capacity challenge and throughput challenge that earlier existed in that area simply by surgically deploying an additional base station or two of the 5G technology. So to summarize, a non-standalone network has the same basis as the 4G network, but it features selective deployment of 5G base stations where there are intense capacity challenges. And in those areas, because phones can leverage two simultaneous channels at the same time, the data rate that they can get in aggregate is substantially higher than the data rate that they can get on the LTE channel. And that solves the throughput and capacity challenge that we were talking about. Now that we have understood the premise of non-standalone networks, let's now try to take a step back and see why they are called non-standalone networks to begin with. Well, as I said, the 5G deployment in this option is merely surgical. You cannot have a nationwide deployment of 5G base stations under non-standalone options. Or even if you don't, you can continue operating your 5G network at least in those select areas. And because in this parlance, the 5G technology cannot stand on its own, there is no end-to-end 5G connectivity. That's why it is called a non-standalone option. 5G connectivity is limited to the wireless channel wherein 5G protocols will continue to operate. But remarkably, this 5G base station in the background will communicate with a 4G core network, not a 5G core network. And that is what I mean by a lack of end-to-end 5G connectivity, although there is no end-to-end 5G connectivity here. One of the major expenses that we discussed in the standalone option is altogether eliminated here. In that 5G base stations will be backward compatible to be able to talk to 4G core network and that is what will enable end-to-end communication from this DC capable UE to the core network, and a beyond core network to the services network. So that is the fundamental premise of non-standalone and standalone options. Standalone option obviously is much cleaner and this is where we'll want to be tomorrow and beyond. But because that is difficult to achieve in the short term, a stepping stone, a middle ground between the two is non-standalone option because of the surgical deployment of 5G base stations. This option is not only more economical, but it is also significantly faster to deploy as compared to standalone networks. So to facilitate 5G services where there is the most demand, operators can start out their deployments with the non-standaloan option. And as they can increase their 5G coverage footprint to a nationwide network, that is when the network will begin to morph into a standalone networks. So NSA is only a stepping stone, a temporary solution. Long term solution is standalone option. And now that we have a good idea about not only the 5G network architecture, but also some of the deployment options. Let's try to scratch the surface on some of the principal techniques that 5G utilizes to keep some of the ambitious promises that it has made and we have previously seen.