Transport Service Provider

Description

The Transport Service Provider (TSP) is a functional concept within the 3GPP service-based architecture, particularly defined for the 5G Core (5GC) network. It acts as a logical intermediary that offers transport connectivity services between various Network Functions (NFs), such as the Access and Mobility Management Function (AMF) and the Session Management Function (SMF), or between User Plane Functions (UPFs). The TSP is not a single physical node but a capability that can be implemented within network functions or as a dedicated service. Its primary role is to manage the transport layer aspects of communication, including the selection of appropriate transport protocols, quality of service (QoS) handling for the transport network, and potentially security aspects like encryption for the N2 (control plane) and N3 (user plane) interfaces.

Architecturally, the TSP interacts with other network functions via service-based interfaces (e.g., Nnssf, Nnef) or reference points. It utilizes transport profiles, which are sets of parameters defining the required transport characteristics (e.g., latency, bandwidth, reliability) for a given service data flow or network slice instance. The TSP consults the Network Slice Selection Function (NSSF) and Policy Control Function (PCF) to determine the appropriate transport resources aligned with the slice's Service Level Agreement (SLA). It then interfaces with the underlying transport network (e.g., managed by a Transport Network Controller) to establish the necessary connectivity, often leveraging technologies like Segment Routing or MPLS.

In operation, when a network function requires transport services, it invokes the TSP's capabilities through standardized APIs. The TSP evaluates the request against available network slice and subscriber policies, selects a transport path, and configures the transport network elements accordingly. This abstraction allows the core network functions to be agnostic of the specific transport technologies (e.g., IP, Ethernet, optical), facilitating multi-vendor interoperability and seamless evolution of the transport layer. The TSP is a key enabler for end-to-end network slicing, as it ensures the transport segment meets the specific performance and isolation requirements of each slice.

Purpose & Motivation

The TSP was introduced to address the growing complexity of managing transport resources in a decoupled, cloud-native 5G core network. In pre-5G architectures, transport connectivity was often tightly integrated with core network nodes, making it difficult to dynamically allocate resources for diverse services like enhanced Mobile Broadband (eMBB), Ultra-Reliable Low-Latency Communications (URLLC), and massive Machine-Type Communications (mMTC). This rigidity hindered efficient network slicing, where each slice requires distinct transport characteristics.

The TSP concept solves this by providing a standardized abstraction layer for transport services. It separates the service logic of the core network from the underlying transport infrastructure, allowing operators to manage and optimize the transport network independently. This enables flexible service deployment, automated provisioning of transport resources per slice, and improved utilization of network assets. Historically, without a TSP, achieving end-to-end QoS and slice isolation across the transport network required proprietary integrations, limiting scalability and innovation.