External Network Slice Information

Description

External Network Slice Information (ENSI) is a critical component of 3GPP's network slicing framework, particularly for roaming scenarios. It is defined as a string that uniquely identifies a network slice instance. When a User Equipment (UE) roams into a visited Public Land Mobile Network (VPLMN), the home network (HPLMN) can include the ENSI within the subscription data or during the registration procedure to indicate which specific slice instance the UE is authorized to use. The VPLMN uses this information, along with other parameters like the Subscribed Network Slice Selection Assistance Information (S-NSSAI), to map the request to a locally available and equivalent Network Slice Instance (NSI). The ENSI is carried within Non-Access Stratum (NAS) signaling messages, such as the Registration Request, and is processed by the Access and Mobility Management Function (AMF) in both the home and visited networks.

Architecturally, ENSI operates within the Network Slice Selection Function (NSSF) and the AMF. The HPLMN's Unified Data Management (UDM) stores the ENSI as part of the user's subscription context. During initial registration or service request procedures, if the UE is roaming, the HPLMN's AMF or NSSF may determine the appropriate ENSI and include it in the response to the VPLMN. The VPLMN's AMF, upon receiving the ENSI, interacts with its local NSSF. The NSSF performs the mapping from the received S-NSSAI and ENSI pair to a VPLMN-specific NSI. This mapping is crucial because the slice identifiers (S-NSSAIs) are only unique within a single PLMN; the ENSI provides the additional granularity needed for the VPLMN to select the exact slice instance intended by the home operator.

The role of ENSI is to bridge the semantic gap between network slice identifiers in different administrative domains. Without ENSI, a VPLMN might only receive an S-NSSAI, which describes a slice type (e.g., enhanced Mobile Broadband, Massive IoT), but not the specific instance. Multiple instances of the same slice type (e.g., different eMBB slices for different enterprise customers) could exist in the HPLMN. ENSI allows the HPLMN to pinpoint the exact instance, enabling the VPLMN to provide a service experience that aligns with the home network's slice policies, including specific Quality of Service (QoS) characteristics, security policies, and network function configurations. It is a key enabler for advanced roaming agreements involving network slicing, ensuring end-to-end slice consistency and fulfilling Service Level Agreements (SLAs) across operator boundaries.

Purpose & Motivation

ENSI was introduced to solve the problem of network slice selection and identification in inter-PLMN (roaming) scenarios. As 5G networks deployed network slicing to offer tailored logical networks for different services, a mechanism was needed to extend these capabilities to users outside their home network. Prior to ENSI, slice selection was primarily based on the S-NSSAI, which is only unique within a single operator's network. During roaming, a visited network receiving only an S-NSSAI would not know which specific instance of that slice type to assign, potentially leading to incorrect service mapping or the inability to honor detailed slice-specific SLAs.

The creation of ENSI was motivated by the commercial need for global service continuity for enterprise and vertical industry applications using network slices. For example, a multinational corporation using a dedicated URLLC slice for factory automation would require that slice's characteristics to be maintained when an employee's device roams onto a partner network. ENSI provides the technical parameter that allows the home network to express, "Use *this specific* URLLC slice instance," enabling the visited network to fulfill that request if a compatible slice is available under the roaming agreement. It addresses the limitation of previous approaches where slice awareness was confined to a single network, thereby unlocking the full economic and technical potential of 5G network slicing in a global ecosystem.