Data Network Access Identifier

Description

The Data Network Access Identifier (DNAI) is a key enabler for edge computing and localized service access in the 5G System. It identifies a specific point where a PDU Session can access a Data Network, which often corresponds to a User Plane Function (UPF) instance deployed at a network edge location close to the user or the application server. The DNAI is not a network address but a logical identifier mapped by the 5G Core to the actual transport and UPF resources needed to reach a particular DN or service instance.

Architecturally, DNAI information is used within the Network Exposure Function (NEF) and the Session Management Function (SMF). An Application Function (AF), such as a video streaming server or an industrial IoT platform, can provide the DNAI to the 5G Core via the NEF to influence PDU Session routing. This is done through the AF influence on traffic routing service defined in TS 23.501. The AF request includes the DNAI and the desired traffic routing policy. The Core Network, specifically the SMF, then uses this DNAI, along with the UE's current location, to select a UPF that provides connectivity to the data network at the point identified by that DNAI. This may involve establishing a new PDU Session Anchor (PSA) UPF or relocating an existing one.

How it works involves close interaction between the application layer and the network layer. When a UE starts an application requiring edge resources, the AF signals the need for a connection via a specific DNAI. The SMF consults the Network Repository Function (NRF) and its local configuration to map the DNAI to a specific UPF instance (or set of instances) that can provide the required access. The SMF then manages the N4 session to the selected UPF and establishes the N6 connection to the DN. This dynamic steering allows the same logical service (identified by a Data Network Name, DNN) to be accessed through different physical points (identified by DNAIs) based on user mobility and application requirements, which is fundamental for ultra-reliable low-latency communication (URLLC) and mobile edge computing (MEC).

Purpose & Motivation

DNAI was created to solve the problem of static, suboptimal traffic routing in mobile networks, which was a major barrier to low-latency services like autonomous driving, industrial automation, and augmented reality. Prior to DNAI, a PDU Session to a Data Network (identified by a DNN) would typically egress the mobile core at a centralized location, adding significant latency for edge-based applications. There was no standardized way for an application to request a specific, geographically optimal access point.

Its introduction in Release 15 was motivated by the 5G requirement to support Edge Computing and network slicing with localized services. The DNAI provides the missing link between the application's awareness of where its service is hosted (e.g., at a specific edge data center) and the network's ability to route the user's traffic to that exact point. It addresses the limitation of the DNN, which only identifies *which* network to connect to, but not *where* to connect from within the mobile network's topology.

This capability unlocks new business models and technical possibilities. It allows operators to deploy multiple instances of the same service across their network and dynamically direct users to the closest one. It is essential for fulfilling the latency and bandwidth promises of 5G for vertical industries, enabling them to treat the mobile network as a distributed compute platform rather than just a connectivity pipe.