Requested Quality of Service

Description

Requested QoS is a fundamental concept in 3GPP networks that defines the quality of service parameters that an application or user requests from the network at the start of a communication session. It is part of the broader QoS framework that ensures different traffic types receive appropriate treatment. The requested QoS is typically expressed as a set of parameters such as maximum bit rate, guaranteed bit rate, delay, jitter, and packet loss rate. These parameters are included in session establishment requests, such as PDP context activation in GPRS/UMTS or PDU session establishment in 5G. The network then evaluates the request against its capabilities, available resources, and subscription policies to decide whether to accept, modify, or reject the requested QoS.

The concept of requested QoS is closely tied to the QoS architecture defined in 3GPP specifications. In UMTS, the QoS profile includes attributes like traffic class (conversational, streaming, interactive, background), maximum bit rate, guaranteed bit rate, transfer delay, and traffic handling priority. In LTE, the QoS is managed through bearers, each associated with a QoS Class Identifier (QCI) and Allocation and Retention Priority (ARP). In 5G, the QoS model uses QoS Flows identified by a QoS Flow Identifier (QFI), with parameters such as 5QI, Guaranteed Flow Bit Rate (GFBR), and Maximum Flow Bit Rate (MFBR). The requested QoS is the initial set of parameters proposed by the UE or the application, which may be modified during the session through QoS modification procedures.

QoS modification requests can occur during the lifetime of a QoS session. These modifications may be triggered by the application (e.g., changing from a video call to a file transfer), by the network (e.g., due to resource constraints or policy changes), or by the user (e.g., upgrading to a premium service). The network must handle these requests while maintaining the agreed QoS for other sessions. The 3GPP standards define procedures for QoS negotiation, authorization, and enforcement, ensuring that the requested QoS is aligned with the subscriber's profile and the operator's policy. The Policy and Charging Control (PCC) framework, introduced in Release 7, plays a key role in authorizing and enforcing QoS requests.

The requested QoS is distinct from the negotiated or granted QoS. The network may not be able to fulfill the exact request due to limitations such as radio conditions, backhaul capacity, or core network load. In such cases, the network may offer a lower QoS, and the application must decide whether to accept it or abort the session. This negotiation is essential for efficient resource utilization and for providing a satisfactory user experience. The 3GPP specifications define the signaling flows for QoS request, negotiation, and modification, ensuring interoperability between different vendors' equipment.

In summary, Requested QoS is a key enabler for differentiated services in mobile networks. It allows applications to specify their requirements, enables operators to manage network resources effectively, and provides a framework for charging based on the quality of service delivered. The concept has evolved from simple PDP context parameters to sophisticated QoS flows in 5G, but the fundamental principle remains: the network strives to meet the requested QoS while balancing the needs of all users.

Purpose & Motivation

The Requested QoS concept exists to provide a standardized way for applications and users to communicate their quality of service requirements to the network. Without it, the network would have no basis for allocating resources appropriately, leading to either over-provisioning (wasting resources) or under-provisioning (causing poor user experience). By allowing the application to request specific QoS parameters at session setup, the network can make informed decisions about admission control, resource reservation, and scheduling. This is particularly important for real-time services like voice and video, which require guaranteed bit rates and low latency.

Historically, early mobile networks like GSM provided a single best-effort service for data. With the introduction of GPRS, basic QoS differentiation became possible through precedence classes, but it was limited. UMTS Release 99 introduced a more comprehensive QoS framework with four traffic classes and a set of parameters, allowing applications to request specific treatment. However, the initial QoS model was complex and rarely used in practice. LTE simplified the model with bearers and QCIs, making QoS more manageable. 5G further refined it with QoS flows and the 5QI table, providing greater flexibility and scalability.

The Requested QoS mechanism addresses the limitations of best-effort service by enabling service differentiation. It allows operators to offer tiered services, charge differently based on QoS, and ensure that critical applications (e.g., emergency services, industrial IoT) receive the necessary resources. It also supports the growing demand for diverse applications with varying requirements, from low-latency industrial control to high-throughput video streaming. The ability to modify QoS during a session ensures adaptability to changing conditions and application needs.