In the previous post in our Techsplanation series, we talked about the radio spectrum and the role it plays in mobile communication networks. In this post, we will talk about some of the ways that mobile network operators maximize their use of spectrum bands. As before, please refer to this glossary for quick reference to some of the key terms and concepts (in bold).
We provide the following information as a foundation for understanding mobile communications generally, but it is especially important for understanding the technologies central to modern and emerging mobile networks like 5G. Again, this is all far too complex to explain in more than fairly abstract terms, but please click through to the linked resources for more detail.
At the end of the last post, we talked about how digital information is transmitted wirelessly through the air and how carriers face many tradeoffs for how to best use the small portions of spectrum they license from the FCC. One of these tradeoffs – between broader coverage with fewer antennas but lower information throughput at lower frequencies and shorter-range coverage with more antennas but higher throughput at higher frequencies – has seen carriers trending toward the latter. Each generation of mobile networks has achieved faster data transmission rates for higher numbers of customers by using higher frequencies than previous generations, but this trend also requires more antennas and transmitters because higher frequencies have shorter effective ranges. Basically, moving up the spectrum toward higher frequencies means more cells, which are smaller in coverage area but are capable of carrying more information, faster.
Ok, I get that, but it doesn’t seem like there is enough spectrum to give each of the nearly 400 million mobile customers in the US their own frequency, so why aren’t our data streams all mixed together?
The spectrum is limited in the sense that there is only one and we cannot produce more of it. However, it is also an inexhaustible resource in that, no matter how many radio signals we send, we can always send more. So lots of super smart people have been figuring out how to get the most out of the spectrum we have. There are essentially four broad categories of techniques for maximizing spectral efficiency. These techniques can be roughly sorted according to whether they separate radio signals based on frequency, time, space, or code.
Each of these techniques can be used for both wireless and wired communications. For instance, frequency division is commonly used in the broadcast radio context – that’s what keeps the classic rock station separate from the pop station on your FM dial. But frequency division is also used in fiber optic communications by separating channels according to frequencies / colors. In the next few paragraphs, we’ll talk about some common forms of frequency and space divisions (those licensed blocks of spectrum allocated by the FCC and NTIA), but carriers employ much more complex methods of squeezing all they can out of their spectrum blocks.
The practical limitations on spectrum usage are that radio signals can get messed up by physical interference (things getting in the way) and they can also interfere with each other. Send too many signals using the same frequency in the same geographical area and it can be tough to sort out which is which. If you’ve ever listened to an FM radio station at the edge of its range, you may have heard competing signals coming from transmitters using the same frequencies. Without some sort of control, the radio spectrum would be a tangled mess of competing signals, and none of them would work very well.
Enter the federal government. The FCC assigns licenses for spectrum usage by dividing the available spectrum into segments (24.75-25.25 GHz) which are subdivided into blocks of varying sizes (in this case, 5 blocks of 100 MHz each). Use of these 100 MHz wide blocks is divided up by geographical area (there are about 416 total licenses available for each block in this segment), and carriers bid for the option to obtain licenses depending on where they wish to operate and what parts of the spectrum are available.
For example, wireless provider x may be licensed to use frequencies between 400 and 450 megahertz in the counties around Orlando, FL. They are not allowed to use frequencies outside that range of spectrum, nor broadcast signals outside that geographical area. This is how we can have different radio stations operating on the same frequencies in different parts of the country. It is also how we can have multiple radio stations and mobile network operators in the same location – each uses a slightly different portion of the spectrum so their signals don’t interfere with each other.
These blocks of spectrum can be further subdivided into channels according to the carrier’s requirements, with wider channels having greater capacity for information throughput. This is where the term bandwidth comes from – the width of the band (the range of adjacent frequencies used) determines how much information you can send (per second). On 3G networks, each connected device uses a channel about 5 MHz wide – say all the frequencies between 400-404.9 MHz.
Ok, but I still don’t see how carriers can connect so many people using only small parts of the spectrum. How many different connections can be made without getting the signals crossed?
Good question. To answer it fully would require several more posts, but we can cover the basics. Even a 100 MHz block would quickly be maxed out if carriers were limited to giving only one connection for each 5 MHz channel. In addition to severely limiting the potential of the spectrum, this would also be terribly inefficient because almost no one is constantly sending and receiving the amount of data that a channel can support. Indeed, many of the advances in mobile communications are based on ways to get more out of the same amount of spectrum.
As I mentioned earlier, carriers can separate wireless signals from each other by dividing and sorting them based on time, space, frequency, and code. In fact, most carriers do all of the above. We’ve already covered the basic ideas of frequency and space division, as administered by the FCC, which amount to cutting the spectrum into blocks and the nation into geographical coverage areas. Carriers do the same thing, but within their allocated spectrum blocks and licensed locations. We won’t dig any further into frequency-division for now, but we will revisit some methods of space-division later in this post.
In time division, many different communication streams can share a common frequency by assigning time slots to each. For instance, Stream A gets to send / receive signals in the first 2 milliseconds of each second, Stream B in the next 2 millisecond interval, etc. The endpoints of these communications synchronize the incoming signals. In reality, these time intervals are probably much smaller than 2ms, and more frequent – otherwise just sending a short message could take multiple seconds – way too long!
Code-division multiple access (CDMA) technology lets carriers use the same channel (same range of frequencies) to serve multiple customers. Essentially, code division creates unique signals for each customer by transmitting two waves that overlap in a special way, allowing the receiving antennas to keep multiple data streams separated. The overlap creates a kind of wave pattern that is distinctive for each of the multiple users sharing a channel. This lets the receiving antennas (both on the user’s device and at the cell tower) filter out all the other signals using the same channel. It’s a bit like having each device speak a different language. They are all audible (to the antennas receiving those frequencies) but it is possible to isolate the language spoken by each.
We have already talked about one form of space division, in which similar frequency blocks are geographically isolated from each other to prevent interference. This takes place at the national level (FCC licenses) and at the carrier level (different cells), but it is also possible to implement space division within cells by keeping track of the locations of each device sending and receiving from the same antenna / transmitter set. This is called space-division multiple access and an important aspect of this technology is the smart antenna. In modern networks, these antennas support the concept of multiple-input, multiple-output (MIMO) and multiple-user MIMO communications. MIMO makes it possible to form even more connections between a single antenna station and many mobile devices. This means that more people can connect in the same area at the same time, or, when there are fewer people, their devices can use multiple channels each for even faster speeds.
There are a few ways to implement MIMO, but they all do the same thing – they create multiple, parallel connections between devices and the antennas thereby multiplying the amount of data transmitted. Some antennas are capable of sending and receiving radio signals to / from precise locations through something called “beam forming.” Essentially, this allows an antenna array to communicate with many different mobile devices on the same frequency by isolating each signal, directionally. The antenna tracks the mobile device’s location and transmits signals meant for a specific device (mostly) only toward that location. Likewise, signals sent from a mobile device will arrive at the antenna at a certain angle, allowing the antenna to differentiate between many different mobile devices using the same swath of spectrum. With MIMO, carriers can get even more use / reuse out of a chunk of spectrum and provide higher data speeds at the same time.
I think I’ve got it—cellular networks use a few different parts of the spectrum—and cell size is a function of frequency and power, and carriers use all kinds of neat tricks to get the most out of their spectrum allocations…but what about the “mobile” part of these communications? What happens when my device moves?
The way cellular networks handle devices moving between cells is so seamless that we rarely think about what is going on in the background. And what’s going on in the background is really complicated. Essentially, your device and the network are communicating with each other (and other parts of the network) about where you are, where you might be going, and how to best support your device’s communication needs while also balancing the needs of other network users. Based on this, your phone and the network figure out which cell or cells are going to provide the best coverage for you as you exit one cell and enter another. Let’s break that down a bit.
As a general matter, your carrier’s network needs to know where your device is so it can successfully route calls and other information to you. Your device helps the network keep track of it by sending “pings” out, wirelessly, which are received by one or more antennas. These pings bounce back to your device (or don’t) which gives the network and your device an idea of how far the device is from the nearest antennas and how strong the signal is between them. The network and your phone can then decide which antenna(s) to use. When your device is farther from the antenna, it must transmit its signal at a higher power to ensure a quality connection. When antennas are closer, devices can transmit at lower power. This is partly why your device’s battery may drain more quickly when there are fewer “bars” in the signal strength indicator icon – it must use more power to keep in touch with the network.
So, unless your device is in “airplane mode” it is intermittently pinging the network so that the network knows how to contact it. In many cases, your device’s pings are registering at multiple antennas if the signals from your phone reach more than one antenna. Based on the strength and timing of these pings, the network can deduce not only where your device is, but also where it is heading. So when you are talking on your phone and driving, your phone is sending and receiving information from the nearest antenna, but is also looking for and talking to the antennas in the next cells. As you approach the border between cells, your phone begins sending two streams of data – you can think of it like sending a copy or duplicating information – to the cell you are currently in and also to the adjoining cell. This duplication of information is a big part of how devices can travel across mobile networks made up of many, many cells without disrupting the information flowing to/from the device. Keep in mind that, in addition to multiple cells, the network may also have multiple channels (remember frequency division?) from which to choose. And remember that this is happening for many, many devices across many, many cells all the time. So, yeah… it’s really complicated.
But wait! There’s more! The next post will cover some of what happens in the core network, because a big part of the transition to 5G involves core network technologies. Don’t worry, it doesn’t get any less complex.