Similar to Evolved Packet System (EPS) in case of 4G, today we will discuss about 5G System (5GS).
5G System includes:
- Core Network a.k.a 5G Core Network (CN) ---- This core network provides connectivity to the internet and to application servers
- Access Network a.k.a 5G Access Network (AN) --- Access network can be a 3GPP Next Generation Radio Access network (NG RAN) or a non-3GPP Access Network
- User Equipment (UE)
The block diagram of 5G System is shown below
5G System (5GS) Block diagram understanding
Let's unpack the above block diagram of 5GS
3GPP Next Generation Radio Access Network (NG RAN):
3GPP NR RAN can be any of the following cases
- New Radio (NR) Base Station. A new radio base station is known as gNodeB, whereas an LTE Base Station which is upgraded to allow connectivity with 5G Core Network --- it is known as an enhanced eNodeB or a Next Generation eNodeB
- Long Term Evolution (LTE) Base Station upgraded to allow connection to the 5G Core Network
- Non-Standalone base station using New radio as the anchor and LTE as an extension
- Non-Standalone Base Station using LTE as the anchor and NR as an extension
Non-3GPP Access Network:
- An example of non-3GPP Access Network is a Wireless Local Area Network (WLAN) based upon WiFi.
- Non-3GPP Access Network use a Non-3GPP Interworking Function (N3IWF) to allow connectivity with the 5G Core Network
- Non-3GPP Interworking Function (N3IWF) supports 3GPP interfaces towards the 5G Core Network and non-3GPP interfaces towards the non-3GPP Access Network
Difference between EPS (4G) and 5G System (5GS)
An important characteristic of the 5G system architecture is the separation of user plane control plane functions.
In 4G, the packet gateway belonging to the original 4G Evolved Packet Core (EPC) provides both control and user plane functions. For example it provides control plane function of IP address allocation, and it provides the user plane function of packet forwarding.
5G system architecture includes the Session Management Function (SMF) for IP address allocation and the User Plane Function (UPF) for packet forwarding. In other words control and user planes are separated.
User and control plane separation allows independant scaling of the two functions. As a use case, operators can add more user plane capacity without having to add control plane capacity.
It also allows different deployment strategies to be adopted for the user plane and control plane. As a use case, user plane functions could be distributed while control plane function could be centralised. Distributing user plane functions helps to keep them place at close locations geographically to maintain minimum latency requirements.
References
- 3GPP TS 23.501
- 3GPP TS 23.214
- 5G NR in Bullets by Chris Johnson