One of the most critical parts of a mobile network is its core network which is also sometimes referred to as the mobile core. Mobile networks are complex and consist of various key entities in their architecture. Depending on the network technology, e.g. GSM, UMTS, LTE, NR, etc., the architecture can look different. These network entities are then grouped to form various parts of the overall mobile network, including the radio network, core network and transport network. The mobile network is then connected to the external networks such as PSTN and Internet to communicate with the outside world. Before we get into the functional details, let’s first define the core network to understand what it does and why it exists?
A mobile core network is a central part of the overall mobile network. It allows mobile subscribers to get access to the services that they are entitled to use, e.g. international calling. The mobile core network is responsible for critical functions such as subscriber profile information, subscriber location, authentication of services and the necessary switching functions for voice and data sessions.
With the evolution of mobile networks from 2G to 3G, 4G and 5G, the mobile core network has also evolved considerably. In the early days of digital mobile communications, the core network for 2G GSM was called Network Switching System (NSS). NSS only supported circuit-switching, which mainly enabled voice calls, SMS and limited data services. Have a look at this post from GSMA to learn more about the history of GSM.
Later, with GPRS (General Packet Radio Service), packet-switching was introduced into the mobile core network to support efficient data services (mobile internet). As a result, two nodes, SGSN (Serving GPRS Support Node) and GGSN (Gateway GPRS Support Node) became part of the 2G core network. The 3G UMTS (Universal Mobile Telecommunication System) core networks followed the same approach for circuit-switched and packet-switched services. The 4G LTE networks (Long Term Evolution), however, use a more advanced core network called Evolved Packet Core or the EPC.
Mobile core network for 2G GSM and 3G UMTS
The mobile core network in the original GSM architecture is known as Network Switching System (NSS). It consists of Mobile Switching Centre (MSC), Home Location Register (HLR), Visitor Location Register (VLR), Authentication Centre (AuC) and Equipment ID Register (EIR). MSC is a fundamental part of the core network which uses other core network components to get subscriber information and enable different services for mobile subscribers. It performs functions such as switching of calls between mobile and fixed users, administration of handovers, authentication and location updates. The original GSM networks used circuit-switched technique for voice calls and SMS, which was supported by the MSC from a core network viewpoint. The GPRS networks employed a packet-switched method and introduced two additional nodes, SGSN and GGSN.
SGSN stands for Serving GPRS Support Node, and GGSN stands for Gateway GPRS Support Node (GGSN), which you can learn more about in this post. SGSN is the packet-switched equivalent of the MSC and works alongside the MSC to enable mobile data services. It is responsible for mobility management, billing, and the management of data sessions. GGSN, on the other hand, sits between the Serving GPRS Support Node (SGSN) and external data networks, e.g. the internet.
When EDGE (Enhanced Data for Global Evolution) was introduced to enhance the existing GPRS networks, it used the same core network architecture. The third-generation UMTS networks also use SGSN, GGSN and MSC in the core network. The high-level network diagram below shows how the core network architecture looks like for a 2G/3G network using GSM and UMTS.
Mobile core network for 4G LTE
4G LTE networks introduced additional nodes into the mobile core network architecture. The LTE core network is called Evolved Packet Core or EPC and uses the packet-switched technique for mobile data as well as voice calls. For packet-switched voice calls, EPC works alongside IMS (IP Multimedia Subsystem) to enable Voice over LTE (VoLTE) calls. VoLTE is also sometimes referred to as 4G calling. It allows you to not only make high-quality voice calls using IP networks but also seamlessly switch between 4G and WiFi calling so you stay connected when the signal quality isn’t great.
If your handset doesn’t support VoLTE, there is a fall back system called Circuit Switched FallBack (CSFB), which allows you to make voice calls using 2G/3G even if you are on 4G. CSFB is also useful if your network doesn’t support VoLTE, e.g. when you are roaming and not entitled to use 4G calling by your operator. More about VoLTE in this dedicated post.
The main network entities within the EPC are Home Subscriber Server (HSS), Mobility Management Entity (MME), Serving Gateway (S-GW), Packet Data Network Gateway (PDN-GW) and Policy & Charging Rules Function (PCRF). These entities are integrated with their 2G/3G counterparts so that inter-technology (e.g. 3G and 4G) handovers and communication can take place. The diagram below provides a high-level view of the 3G/4G core network architecture.
What about 5G core network?
In 5G networks, the core network architecture depends on the kind of 5G deployment. 5G can be deployed in two ways, i.e. standalone and non-standalone, which you can read more about in this post. Non-standalone is currently the most common mode of deployment which uses a new 5G radio network (NR= New Radio), and an enhanced version of the LTE core network (EPC). This approach allows mobile operators to make the most of their existing LTE network infrastructure. The standalone mode, on the other hand, is not as common yet but is expected to be the real game-changer. The standalone deployment will use a new 5G core network (5G Core) alongside a new 5G radio network (NR).
5G Core is a future-proof solution that makes use of cloud-based technologies optimised for cloud-native applications. The advanced use cases for 5G, e.g. critical IoT services, require ultra-low latency, which 5G Core network will be able to support. With a cloud-native core network, the introduction of new services or functionalities can be quicker and more in line with the agile methodologies.