Work series: Cellular Network Basics and Mobility (Countries, visas and border checks)
I decided to give details about the research work we are doing in Nokia but for this I need to share some basics first. This blog aims to start that. Happy Olympics to everyone!
Inspired by Vikram and his successful blog, I finally decided to talk about the topics I am working on as it might raise a higher interest from readers.
I am working in Nokia Standards Research to advance the wireless network performance, specifically on “mobility” topic. If you do not know what is mobility, it will come. Without further ado let us get to the basics.
In this blog, you will get to know (1) what are the cells in cellular networks, (2) what 5G has different compared to previous generations (beams) and (3) why it is challenging to keep a mobile communication system running.
Cells in a cellular network
Let us start with the fundamentals and something you know from the real world.
Aerial view of a base station (Source)
You have all seen towers like the one in the photo. These towers and the necessary radio equipment to operate the radios on the towers are called base stations. Depending on each generation you may have seen terms as BTS (2G), NodeB (3G) eNB (4G) and gNB (5G). Let us refer to them as base stations. Let us abstract this concept with a triangle to avoid using real world photos. And let us abstract the coverage each antenna provides (see below depicted with blue arrows) with a representative hexagon. The mobile phones, referred to as user equipment (UE) can be abstracted with a red dot.
You can subscribe below to complain for image quality for free.
Typically a base station is equipped with multiple antennas as you see in the previous photo. Each antenna can send different signals on different or same radio frequencies. For the sake of simplicity let us consider a same frequency network as it is mainly deployed in the field after 4G (thanks to better interference management techniques.). The area that can receive the signal from each antenna is called a “cell”. And voila you have your cellular part of cellular communications. I will refer to the term network, when I have in mind an entity that controls the base station from now on.
You might ask “ Why each base station is not a single cell?”. The three cell concept allows each antenna to some-extend run “independently” from others. In other words, user equipment (UE, depicted as red dot) needs a specific configuration to communicate to the antenna of the base station, except the border of two cells, the configuration of a single antenna is sufficient. This results in the definition of cell configuration as such the “cell” concept. However, making the a base station a single cell is theoretically possible. Moreover, making the whole network a single cell is also possible. But more on that later.
A base station is also called a site. A site with three sectors is a common combination you can have.
In previous generations the cell concept helped frequency planning where each cell would have a different frequency and the frequencies are repeated in cells of the network with a pattern (e.g., repeat after every 7 cell). In newer generations we do not have this anymore.
Beams of 5G
(I hope in your browser you can still see the above photo.) So when a user equipment (i.e., mobile phone, red dot) connects to a cell, it is actually connecting to a base station (green triangle). A base station controls a cell or gives the cell configuration for the antenna of the cell that UE is communicating to.
Is the coverage in the network organized on the level of cells only? Glad you asked. No, let us go one more step granular, to beams.
5G networks introduced a new frequency band to be used for communications, milimeter waves (mmWave). The introduction of mmWaves enabled (or made a necessity) use of beamforming aka Multiple Input Multiple Output (MIMO) techniques to re-use the same frequency more granular than at a cell level. Imagine that you build 15 cells out of one cell! You are very right. (On the other hand in the market mmWave still did not materialize too much but please do not focus there, we are at a technical level here! )
Your next question would be “Does this mean each beam is a new cell?”. Even though that is a technical possibility the answer is no. Each beam under the same cell is still transmitted by the same antenna, so the configuration is mostly similar except the beamforming configuration (Transmission Configuration Indicator (TCI) configuration). The beamforming configuration is part of cell configuration provided by the base station.
An analogy is in order to remember things well. You can think of cells as different countries and beams as cities within the countries. If we want to exaggerate we can even think of base stations as continents. Let us consider phones as citizens in this global system.
We can do a wrap up now:
In cellular networks, there are towers, that are called base stations. (Continents)
Base stations control cells that depict the coverage area. (Countries)
Through beamforming cells can be composed of beams. (Cities)
Mobility of mobile communications
Why did we learn this? Not because it is interesting to know, which it is, but also to build up to the problems of the cellular networks. In the cellular networks UEs can be in two states, namely Connected and Idle State (three actually but let us ignore the Inactive state).
In the idle state the phone is not connected to any base station. (Extending on the previous analogy these are citizens without countries.) So the network (or your network provider like Vodafone) is not actively working with your phone. However, in connected state the network has a lot of responsibility to provide the best connectivity (up to the bill or subscription you are paying). Let us skip the story of how UE goes into connected state for now. (Story of another post. )
In connected state UE is connected to a cell! “But if I get on a bus while in a phone call, then what happens? The same cell cannot keep providing coverage to me.”. Yes, the network has to make sure that the UE has the correct cell configuration and for this reason cell configuration needs to change.
This procedure is called handover where the cell of the UE is changed. (Further on the analogy, each time a citizen is trying to move between countries there is a border check and a visa needs to be granted so that the citizen can move without an issue. The procedure is literally called “admission control” in standardisation.) There are multiple procedures to make sure that UE is connected to the network.
The above figure is an attempt to summarize these procedures. In order to keep it simple, we will use some categories. We learned about beams and cells. The left yellow box covers the procedures for maintaining the correct beam configuration within the cell. The right orange box covers the procedures for maintaining the correct cell configuration.
Within each procedure you have procedures that happen within the UE depicted with blue boxes. These procedures are failure recovery procedures. These are fallback solutions in case the network cannot react in time. ( It is like a citizen going to an embassy to declare your passport is lost and they should check their records to route the citizen to the right country and city.)
The rest of the procedures are procedures built for network to re-configure the UE in time to provide the right configuration. All of these procedures make use of measurements UE does that are based on signals that the base stations send to the UE. UE sends the measurements to the network and the network has to delete the old configuration and provide the new configuration exactly at the cell or beam border.
In the above example the mobility procedure is illustrated with signal strength. At left hand side figure UE is connected to Cell 12 with a good signal level, the closer to the cell 11 the UE gets, the stronger the Cell 11 becomes. If Cell 11 is too strong the signal of Cell 12 is not decodable anymore. So the network has to detect that UE is moving towards Cell 11(through measurements it sends) and send the configuration of Cell 11 over the connection that UE has with Cell 12.
Now I see this might be confusing. Why does the network has to send the configuration of Cell 11 over Cell 12? Why cannot the UE connect directly to Cell 11? The short answer is it can, but it will take a long time and if there is a call on-going this will result in a drop or a long “Can you hear me?” moment. Long answer is in order to keep your service as smooth as possible, the network has to set up everything to serve you in the next cell. This procedure is called “handover” and the general umbrella is called mobility.
For this week let us leave it at that and next week we can dive in details of what 5G has introduced such as
conditional handover (CHO) that resolves the issue of providing the configuration exactly on cell change
A strong passport with visas granted to multiple countries at once.
dual active protocol stack (DAPS) that enables almost no interruption time during handover
Right of dual citizenship
and layer 1 / layer 2 triggered mobility (LTM) that enables a more cloud RAN aware, beam aware mobility procedure!
European Union citizenship (has its advantages but it is complicated)
Insightful article