Reflective QoS Attribute

Description

The Reflective QoS Attribute (RQA) is a fundamental parameter within the 5G Quality of Service (QoS) framework, specifically designed to support the Reflective QoS mechanism. It is defined as a QoS attribute associated with a QoS Flow. The RQA is provisioned by the Session Management Function (SMF) and is sent to the User Equipment (UE) within the PDU Session Establishment Accept message (or a subsequent PDU Session Modification message) as part of the QoS rule(s) for a specific QoS Flow. The presence of the RQA for a given QoS Flow explicitly instructs the UE that this flow is configured for Reflective QoS operation.

Architecturally, the RQA is a control-plane element that resides within the QoS profile managed by the SMF. When the SMF determines that a QoS Flow should utilize Reflective QoS—often for applications with symmetric uplink and downlink requirements like Voice over IP (VoIP) or real-time interactive services—it includes the RQA in the QoS rule sent to the UE via the Access and Mobility Management Function (AMF). The UE's 5G QoS (5QI) characteristics for the flow are already known, but the RQA provides the specific authorization to apply the reflective derivation process.

How it works is intrinsically linked to the Reflective QoS Indication (RQI). When a downlink packet for a QoS Flow marked with RQA arrives at the UE, the UPF (User Plane Function) or gNB may set the RQI bit in the packet's encapsulation header (e.g., in the GTP-U extension header or the QFI field). The UE's user plane detects this RQI-marked packet. The presence of the RQI in the user plane, combined with the prior authorization signaled by the RQA in the control plane, triggers the UE to create or update an uplink QoS rule. This rule mirrors the QoS characteristics (such as 5QI, ARP, session-AMBR) of the downlink flow, effectively deriving uplink policy from observed downlink traffic without continuous core network signaling.

Its role in the network is to enable efficient, dynamic, and application-aware QoS management. It reduces signaling overhead between the UE and the core network (specifically the SMF) by automating the setup of symmetric QoS flows. This is crucial for supporting low-latency services and network slicing, where rapid adaptation to application needs is required. The RQA acts as the foundational permission slip, while the RQI provides the real-time trigger, together forming a closed-loop, reflective control system for user-plane QoS.

Purpose & Motivation

The RQA was created to address the inefficiencies of traditional, purely network-provisioned QoS models in handling dynamic, symmetric traffic patterns. Prior to 5G, uplink QoS rules were typically established explicitly via control-plane signaling from the network (e.g., PCC rules from the PCRF/PCF). For applications like conversational video or gaming, where uplink and downlink requirements emerge simultaneously with a session, this could introduce setup delay and generate significant signaling load, especially with frequent flow additions or modifications.

The motivation for RQA and the broader Reflective QoS concept stems from the need for faster and more scalable QoS activation. The 5G system aims to support a massive number of devices and diverse service requirements, including those defined for Ultra-Reliable Low-Latency Communication (URLLC). Explicit signaling for every symmetric flow would become a bottleneck. Reflective QoS, authorized by the RQA, allows the UE to act autonomously based on network hints (the RQI), dramatically reducing latency for flow setup and lowering control-plane load.

Historically, QoS was entirely network-controlled. The introduction of RQA in 3GPP Release 15 represents a paradigm shift towards a more distributed intelligence model. It empowers the UE with limited, network-authorized decision-making capability for QoS. This solves the problem of rapid response to application demands and is a key enabler for network slicing, where different slices may employ different QoS strategies, including reflective methods for efficiency.