Reference point between the NG-RAN and the AMF

Description

The N2 interface is a conceptual reference point defined in 3GPP 5G system architecture between the Next Generation Radio Access Network (NG-RAN) node—which includes gNBs (5G base stations) and ng-eNBs (upgraded LTE base stations)—and the Access and Mobility Management Function (AMF) within the 5G Core (5GC). It is a logical point-to-point interface that carries NGAP (Next Generation Application Protocol) signaling messages. NGAP is the application layer protocol that operates over SCTP (Stream Control Transmission Protocol) transport, providing reliable, connection-oriented signaling. The N2 interface is exclusively a control-plane interface; it does not carry user data, which is handled by the N3 interface between NG-RAN and UPF.

Architecturally, N2 is a key component of the service-based architecture (SBA) of 5GC, representing the interaction between the RAN and the core network control plane. Each NG-RAN node establishes one or more SCTP associations with an AMF for redundancy and load balancing. The AMF may serve multiple NG-RAN nodes, and an NG-RAN node may connect to multiple AMFs for resilience. The N2 interface enables functions such as UE registration and deregistration, connection establishment and release, handover signaling, paging coordination, and NAS (Non-Access Stratum) transport between UE and AMF. It also supports network slicing by allowing the NG-RAN to route signaling to the appropriate AMF based on the slice selection.

How N2 works involves a series of NGAP procedures. When a UE attaches to the network, the NG-RAN node selects an AMF based on configuration or information from the UE (like GUAMI—Globally Unique AMF Identifier) and initiates an N2 connection. The NG-RAN sends an NGAP Initial UE Message containing the UE's NAS registration request to the AMF. Subsequent mobility events, like handovers, trigger NGAP Handover Required messages from source NG-RAN to AMF, which coordinates with target NG-RAN via N2. The interface also transports downlink NAS messages from AMF to UE, encapsulated in NGAP Downlink NAS Transport messages. All signaling is secured using IPsec as per 3GPP security specifications.

The N2 interface plays a vital role in enabling 5G's flexibility and scalability. By separating the control plane (N2) from the user plane (N3), it allows independent scaling of control and data processing functions. It supports stateless AMF operations, where the AMF can be changed without losing UE context, by storing context separately. N2 also facilitates edge computing scenarios by allowing local AMF deployments. In network slicing, N2 carries the S-NSSAI (Single Network Slice Selection Assistance Information) to ensure the UE is served within the correct slice. The interface is defined to work over any IP transport network, accommodating various deployment options including centralized and distributed RAN architectures.

Purpose & Motivation

The N2 interface was created as part of the 5G system architecture to address limitations in previous generations, particularly the tight coupling between control and user plane in LTE's S1 interface. In LTE, the S1 interface connected eNB to both MME (control plane) and S-GW (user plane), which constrained flexibility and innovation. The 5G design principles demanded a clear separation of control and user plane, leading to the definition of distinct N2 (control) and N3 (user) interfaces.

Its primary purpose is to provide a standardized, flexible signaling path between RAN and core for access and mobility management, enabling features like network slicing, edge computing, and service-based core architecture. N2 solves the problem of efficiently managing mobility and sessions in a heterogeneous network with mixed 4G/5G access, as it supports both gNBs and ng-eNBs. It also allows for AMF redundancy and load balancing, improving reliability compared to LTE's single MME connection per eNB.

Historically, the evolution from GTP-C (GPRS Tunneling Protocol for control plane) in 4G to NGAP in 5G reflects a shift towards more modular, cloud-native networks. N2 enables stateless operation, where AMF instances can be added or removed dynamically, supporting cloud deployment models. It was motivated by the need for lower latency signaling, better support for massive IoT, and the ability to integrate with multi-access edge computing (MEC). By providing a clean interface, N2 allows RAN and core networks from different vendors to interoperate, fostering an open ecosystem.