Service Data Adaptation Protocol

Description

The Service Data Adaptation Protocol (SDAP) is a sublayer within the 5G New Radio (NR) user plane protocol stack, residing above the Packet Data Convergence Protocol (PDCP) and below the application layer. It operates transparently between the gNB (Next Generation NodeB) and the User Equipment (UE). Its primary architectural role is to act as an adaptation layer for the QoS framework defined for the 5G System (5GS). Unlike in LTE, where QoS was tied to EPS bearers, 5G introduces a more flexible QoS model based on QoS Flows. The SDAP entity is configured per Protocol Data Unit (PDU) Session and per data radio bearer (DRB).

SDAP works by processing downlink packets from the core network's User Plane Function (UPF) and uplink packets from the UE's higher layers. Each IP packet is associated with a specific QoS Flow Identifier (QFI). The core function of SDAP is to map these QoS flows onto the appropriate data radio bearers. A single DRB can carry packets from multiple QoS flows if they share similar QoS characteristics, a process known as QoS flow aggregation. Conversely, a QoS flow with stringent requirements might be mapped to a dedicated DRB. The SDAP entity in the gNB performs this mapping in the downlink, while the UE's SDAP entity performs the reverse mapping in the uplink based on rules received from the network.

A key operational mechanism is the marking of packets with QFI headers. In the downlink, the gNB's SDAP layer adds a small SDAP header to the packet, which includes the QFI and potentially an RQI (Reflective QoS Indicator) and/or an RDI (Reflective QoS Indication for Delay Critical GBR). This header allows the UE to identify the QoS flow to which the packet belongs for proper uplink treatment. For reflective QoS, the RQI bit instructs the UE to create a mirroring QoS rule for the uplink based on the observed downlink traffic, reducing signaling overhead. The SDAP layer is also responsible for handling the establishment, modification, and release of SDAP entities and their associated mappings via RRC signaling.

Purpose & Motivation

SDAP was created to support the revolutionary QoS model of the 5G System, which was designed to cater to an unprecedented variety of services—from enhanced mobile broadband (eMBB) to ultra-reliable low-latency communications (URLLC) and massive IoT (mIoT). The previous 4G EPS bearer model was relatively rigid, binding QoS parameters to a bearer tunnel end-to-end. This made dynamic service creation and fine-grained traffic differentiation cumbersome. The 5G QoS model decouples the QoS flow (a service-level concept) from the data radio bearer (a transport-level concept), enabling greater flexibility and efficiency.

The protocol exists to solve the problem of efficiently mapping these abstract QoS flows onto the physical radio resources (DRBs) while maintaining the integrity of QoS enforcement. It allows the network to optimize radio resource usage by aggregating multiple similar flows onto one bearer or isolating critical flows on dedicated bearers without requiring core network involvement for every adjustment. Furthermore, SDAP enables network slicing by providing a clear demarcation point where slice-specific QoS policies, received from the core network, can be applied to the radio bearer mapping. Its introduction was motivated by the need for a protocol mechanism that could realize the advanced 5G QoS framework, ensuring that diverse latency, reliability, and bandwidth promises could be technically enforced on the air interface.