Peak Bit Rate

Description

Peak bit rate (PBR) is a fundamental QoS parameter defined in 3GPP TS 21.905 and TS 22.060. It represents the maximum bit rate that the network can offer to a user for a specific time period, typically for the transfer of a bursty signal. The definition is derived from ITU-T Recommendation I.210, which describes it as a measure of throughput and the maximum user information transfer rate achievable for a single service data unit transfer. In essence, PBR is the highest instantaneous rate at which data can be delivered to or from a user, and it is used to characterize the burstiness of traffic and to ensure that the network can handle short-term peaks without violating QoS guarantees.

In the context of 3GPP networks, peak bit rate is closely related to other QoS parameters such as Guaranteed Bit Rate (GBR), Maximum Bit Rate (MBR), and Aggregate Maximum Bit Rate (AMBR). While GBR ensures a minimum bit rate for a bearer or flow, PBR specifies the upper limit that should not be exceeded, except perhaps for very short durations. For GBR bearers, the MBR is typically set equal to the GBR, but PBR can be used to allow some burstiness above the GBR, as long as the average rate remains within the GBR. In non-GBR bearers, the MBR and AMBR are used to cap the maximum rate, and PBR may be used in conjunction with these to define the peak allowed for short bursts.

The peak bit rate is particularly important for traffic shaping and policing functions in the network. The UPF and the RAN employ traffic policing to enforce the PBR, dropping or marking packets that exceed the peak rate. This prevents a single user or application from consuming excessive resources and ensures fairness among users. The PBR is also used in admission control and resource reservation: when a new flow is requested, the network checks whether the peak bit rate can be supported without degrading existing flows. If not, the request may be rejected or the flow may be downgraded.

The time period over which the peak bit rate is measured is not strictly defined in the standard, but it is typically on the order of milliseconds to seconds. This allows the network to accommodate bursts that occur on short timescales while still enforcing an average limit. For example, a video streaming application may have an average bit rate of 5 Mbps but bursts up to 10 Mbps for a few hundred milliseconds when a new frame is loaded. The PBR would be set to 10 Mbps to allow these bursts, while the GBR might be 5 Mbps to ensure consistent quality.

In 5G, peak bit rate remains a relevant parameter for QoS enforcement, particularly for eMBB and URLLC services. The 5G QoS model uses 5QI values that define default characteristics, including the maximum data burst volume (MDBV) and the packet delay budget (PDB). The PBR is not explicitly part of the 5QI definition but can be configured per QoS Flow via the SMF. The SMF communicates the PBR to the UPF and RAN using N4 and N2 interfaces, respectively. In Release 19, the concept of PDU Sets further refines QoS enforcement, and PBR may be applied to PDU Sets to limit the burstiness of application-level units.

Overall, peak bit rate is a key tool for network operators to balance service quality with resource efficiency. It allows them to define the maximum rate that a user can achieve, while accommodating the bursty nature of real-world traffic. It is essential for dimensioning, admission control, and traffic shaping in both 4G and 5G networks.

Purpose & Motivation

Peak bit rate was introduced to address the need for a QoS parameter that captures the bursty nature of packet-switched traffic. Early networks, such as circuit-switched systems, had fixed bandwidth allocations, but packet-switched networks like GPRS and UMTS introduced variability in data rates. The ITU-T I.210 definition provided a standardized way to describe the maximum rate for bursty signals, which was then adopted by 3GPP in Release 4.

The primary problem PBR solves is the efficient allocation of network resources while ensuring QoS for bursty applications. Without a peak rate parameter, operators would either have to over-provision resources to handle worst-case bursts (which is costly) or risk violating QoS guarantees during bursts. PBR allows operators to specify a maximum rate that the network can police, ensuring that no single user or application monopolizes resources. It also enables admission control algorithms to determine whether a new flow can be admitted based on its peak requirements.

Furthermore, PBR is essential for traffic shaping and policing, which are critical for maintaining fairness and preventing congestion collapse. By enforcing PBR, the network can smooth out bursts and ensure that the average rate stays within contracted limits. This is particularly important in shared access networks where multiple users compete for bandwidth. In 5G, PBR continues to play a role in QoS management, especially for services that require high reliability and low latency, where bursty traffic must be carefully controlled to meet stringent performance targets.