One of the problems of mobile networks is that the terminal can be anywhere at any given moment. How to reach the terminal, send it a packet wherever it is, without the terminal consuming too much energy by a frequent exchange of messages? That is the question we are going to answer in this video. We know that a mobile network is composed of a set of cells. Base stations are set up in a regular pattern on the territory to offer good coverage. In 4G, a base station is called an eNodeB. We’ll look at a very simple model with square cells that will enable us to make a few simple calculations in order to understand the main phenomena. The patterns we’ll uncover are valid for more complex models, for example, with a model where we use hexagonal cells, which is what is usually used in the literature. The terminal, as I said, can be anywhere at any time, in any cell. If someone wants to reach it, how do we establish a link to wherever it is? Well, the first idea that comes to mind, is to consider the principle of updating location: if we consider a terminal that’s on the move, each time it changes cells, it could send a message to the network, saying, “Now, I’m in this cell, send my packets to this eNodeB.” If we look a bit at what this entails, the first thing is that the terminal must be able to detect a change of cells. To do this, we have the concept of a beacon channel. What is a beacon channel also called Broadcast Control Channel? It’s that each eNodeB broadcasts its identity regularly on a radio frequency, typically every one to five seconds. We have, for example here, enodeB B1 which broadcasts its identity B1, B2 for the other and B3. This way, a cell phone that goes from one cell to another just needs to listen to the beacon channels: when it receives a signal from B2 stronger than from B1, that indicates it has changed cells. Now, let’s look at how many location updates are made in a simple case. We know that, in urban zones, the operator deploys a large number of base stations per square kilometer, which means that each cell is small. We could, for example, imagine that in a very dense zone, a cell is six hundred meters long. If we take a pedestrian who is moving at 3.6 km/h, that comes down to a speed of 1 m/s and he will cover the distance of 600 meters in 600 seconds. Now, let’s suppose that he’s in his car. Now, the speed will be closer to 36 km/h, or 10 m/s. At this point, six hundred meters is covered in 60 seconds. That means that, every minute, the terminal has to update its location, since, on average, it changes cells every minute. If, every minute, even if I’m not using my terminal, it updates its location, it’s going to transmit and receive messages from the base station and it is going to consume energy even though I’m not doing anything. At the end of the day, my battery will be dead, perhaps before the end of the day. So, we’re going to try to reduce the frequency of the location updates. To do so, we’ll group some cells in what is called a Tracking Area or TA. It was known as a location area in 2G or 3G. Now, each base station, instead of broadcasting its identity broadcasts the identity of the tracking area. This identity is made up of a Mobile Country Code (MCC), a Mobile Network Code (MNC), and a Tracking Area Code, a code allocated by the operator. How does this work? The terminal still listens to the beacon channel, if it’s coming for example from base station B10, when it enters zone TA1, it will detect a base station broadcasting a code that is different from the previous one. The terminal detects the change of tracking area. It will update its location, saying, “I’m in TA1,” and not “I’m in cell X.” This means that, when the cell phone moves and goes into a different cell, well, it does nothing, it doesn’t send a signaling message. Now, it can go anywhere in the entire tracking area, it’s always the same, nothing happens. As long as the cell phone does not cross the boundary of the tracking area, it does not update its location. Let’s look at what that enables us to do. We no longer have areas that are six hundred meters long, we now have zones that are 1800 meters long. We take terminals that move in a straight line: instead of updating every minute, we have an update every three minutes, since in three minutes we will have covered, for a terminal, at 10 meters a second, the 1800 meters. We have 9 cells in the tracking area and we’ve divided the average number of location updates that we do by 3, which is the square root of 9. You can see that, now, the network no longer knows in which cell the terminal is located at each time. It only knows the tracking area. That means that, if there was an update at a given moment, maybe at the moment you want to reach the terminal, it is in a different cell. So, when you try to reach the terminal, you have to broadcast its identity to all the cells in the tracking area. Here, you can see that we have multiplied by nine the number of messages that are sent to try to reach a terminal. The principle of broadcasting the same message to several cells is called Paging. Paging consists in broadcasting the identity (the TMSI, Temporary Mobile Subscriber Identity) to all the cells of the tracking area. In conclusion, a tracking area regroups several cells. A UE can move within the tracking area without updating its location. The larger the area, the fewer location updates we have to do. However, there’ll be more paging messages to send. At first approximation, a tracking area of N cells allows you to divide the number of updates made per unit of time by the root of N. The number paging messages is multiplied by N.
