Content for  TS 23.203  Word version:  19.2.0

The service level (i.e. per SDF or per SDF aggregate) QoS parameters are QCI, ARP, GBR and MBR. Each Service Data Flow (SDF) is associated with one and only one QoS Class Identifier (QCI). For the same IP-CAN session multiple SDFs with the same QCI and ARP can be treated as a single traffic aggregate which is referred to as an SDF aggregate. An SDF is a special case of an SDF aggregate. The QCI is scalar that is used as a reference to node specific parameters that control packet forwarding treatment (e.g. scheduling weights, admission thresholds, queue management thresholds, link layer protocol configuration, etc.) and that have been pre-configured by the operator owning the node (e.g. eNodeB). When required by operator policy, the eNodeB can be configured to also use the ARP priority level in addition to the QCI to control the packet forwarding treatment in the eNodeB for SDFs having high priority ARPs.

This clause specifies standardized characteristics associated with standardized QCI values. The characteristics describe the packet forwarding treatment that an SDF aggregate receives edge-to-edge between the UE and the PCEF (see Figure 6.1.7-1) in terms of the following performance characteristics:

1. Resource Type (GBR or Non-GBR);
2. Priority;
3. Packet Delay Budget;
4. Packet Error Loss Rate;
5. Maximum Data Burst Volume (for some GBR QCIs);
6. Data Rate Averaging Window (for some GBR QCIs).

The standardized characteristics are not signalled on any interface. They should be understood as guidelines for the pre-configuration of node specific parameters for each QCI. The goal of standardizing a QCI with corresponding characteristics is to ensure that applications / services mapped to that QCI receive the same minimum level of QoS in multi-vendor network deployments and in case of roaming. A standardized QCI and corresponding characteristics is independent of the UE's current access (3GPP or Non-3GPP). The one-to-one mapping of standardized QCI values to standardized characteristics is captured in Table 6.1.7-A and Table 6.1.7-B. The main differences between the two parts are that, in contrast to Part A, Part B of Table 6.1.7 describes QCIs for which the Packet Error Loss Rate calculation includes those packets that are not delivered within the Packet Delay Budget; and, it provides additional information on the Data Rate Averaging Window as well as the Maximum Data Burst Volume that needs to be delivered within the Packet Delay Budget.

The Resource Type determines if dedicated network resources related to a service or bearer level Guaranteed Bit Rate (GBR) value are permanently allocated (e.g. by an admission control function in a radio base station). GBR SDF aggregates are therefore typically authorized "on demand" which requires dynamic policy and charging control. A Non GBR SDF aggregate may be pre-authorized through static policy and charging control. The Maximum Data Burst Volume, if defined for the QCI (see Table 6.1.7-B), is the amount of data which the RAN is expected to deliver within the part of the Packet Delay Budget allocated to the link between the UE and the radio base station as long as the data is within the GBR requirements. If more data is transmitted from the application, delivery within the Packet Delay Budget cannot be guaranteed for packets exceeding the Maximum Data Burst Volume or GBR requirements. The Data Rate Averaging Window, if defined for the QCI (see Table 6.1.7-B), is the 'sliding window' duration over which the GBR and MBR for a GBR SDF aggregate shall be calculated (e.g. in the RAN, PDN-GW and UE). The Packet Delay Budget (PDB) defines an upper bound for the time that a packet may be delayed between the UE and the PCEF. For a certain QCI the value of the PDB is the same in uplink and downlink. The purpose of the PDB is to support the configuration of scheduling and link layer functions (e.g. the setting of scheduling priority weights and HARQ target operating points). Except for QCIs 82 and 83, the PDB shall be interpreted as a maximum delay with a confidence level of 98 percent. For services using QCI 82 or 83, a packet delayed by more than the PDB is included in the calculation of the PELR if the packet is within the Maximum Data Burst Volume and GBR requirements. The support for SRVCC requires QCI=1 only be used for IMS speech sessions in accordance to TS 23.216. Services using a Non-GBR QCI should be prepared to experience congestion related packet drops and, except for QCI 80, 98 percent of the packets that have not been dropped due to congestion should not experience a delay exceeding the QCI's PDB. This may for example occur during traffic load peaks or when the UE becomes coverage limited. See Annex J for details. Packets that have not been dropped due to congestion may still be subject to non congestion related packet losses (see PELR below). Owing to its low latency objective, services using QCI 80 should anticipate that more than 2 percent of packets might exceed the PDB of QCI 80. Except for services using QCI 82 or 83 services using a GBR QCI and sending at a rate smaller than or equal to GBR can in general assume that congestion related packet drops will not occur and 98 percent of the packets shall not experience a delay exceeding the QCI's PDB. Exceptions (e.g. transient link outages) can always occur in a radio access system which may then lead to congestion related packet drops even for services using a GBR QCI and sending at a rate smaller than or equal to GBR. Packets that have not been dropped due to congestion may still be subject to non congestion related packet losses (see PELR below). For services using QCI 82 or 83 a packet which is delayed by more than the PDB but is within the Maximum Data Burst Volume and GBR requirements, is counted as lost when calculating the PELR. Every QCI (GBR and Non-GBR) is associated with a Priority level (see Table 6.1.7). The lowest Priority level value corresponds to the highest Priority. The Priority levels shall be used to differentiate between SDF aggregates of the same UE and it shall also be used to differentiate between SDF aggregates from different UEs. Via its QCI an SDF aggregate is associated with a Priority level and a PDB. Scheduling between different SDF aggregates shall primarily be based on the PDB. If the target set by the PDB can no longer be met for one or more SDF aggregate(s) across all UEs that have sufficient radio channel quality then the QCI Priority level shall be used as follows: in this case a scheduler shall meet the PDB of an SDF aggregate on QCI Priority level N in preference to meeting the PDB of SDF aggregates on next QCI Priority level greater than N, until the priority N SDF aggregate's GBR (in case of a GBR SDF aggregate) has been satisfied. Other aspects related to the treatment of traffic exceeding an SDF aggregate's GBR are out of scope of this specification. When required by operator policy, the eNodeB can be configured to use the ARP priority level in addition to the QCI priority level to determine the relative priority of the SDFs in meeting the PDB of an SDF aggregate. This configuration applies only for high priority ARPs as defined in clause 6.1.7.3. The Packet Error Loss Rate (PELR) defines an upper bound for the rate of SDUs (e.g. IP packets) that have been processed by the sender of a link layer protocol (e.g. RLC in E-UTRAN) but that are not successfully delivered by the corresponding receiver to the upper layer (e.g. PDCP in E-UTRAN). Thus, the PELR defines an upper bound for a rate of non congestion related packet losses. The purpose of the PELR is to allow for appropriate link layer protocol configurations (e.g. RLC and HARQ in E-UTRAN). For a certain QCI the value of the PELR is the same in uplink and downlink.

The QoS parameter ARP contains information about the priority level, the pre-emption capability and the pre-emption vulnerability. The priority level defines the relative importance of a resource request. This allows deciding whether a bearer establishment or modification request can be accepted or needs to be rejected in case of resource limitations (typically used for admission control of GBR traffic). It can also be used to decide which existing bearers to pre-empt during resource limitations. The range of the ARP priority level is 1 to 15 with 1 as the highest level of priority. The pre-emption capability information defines whether a service data flow can get resources that were already assigned to another service data flow with a lower priority level. The pre-emption vulnerability information defines whether a service data flow can lose the resources assigned to it in order to admit a service data flow with higher priority level. The pre-emption capability and the pre-emption vulnerability can be either set to 'yes' or 'no'. The ARP priority levels 1-8 should only be assigned to resources for services that are authorized to receive prioritized treatment within an operator domain (i.e. that are authorized by the serving network). The ARP priority levels 9-15 may be assigned to resources that are authorized by the home network and thus applicable when a UE is roaming.