0% found this document useful (0 votes)

4 views88 pages

Definition of Quality of Service Parameters and Their Computation 27 January 2025

The GSMA document IR.42 outlines the definitions and computations of Quality of Service (QoS) parameters for telecommunications. It includes a comprehensive model of service-independent and service-specific QoS parameters, detailing metrics for various services such as telephony, SMS, and data services. This document serves as an industry specification and is subject to copyright and compliance regulations.

Uploaded by

Abdo

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

4 views88 pages

Definition of Quality of Service Parameters and Their Computation 27 January 2025

Uploaded by

Abdo

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 88

GSMA

Official Document IR.42 Definition of Quality of Service parameters and their computation

Definition of Quality of Service parameters and their

computation
Version 11.0
27 January 2025

Security Classification: Non-Confidential

Access to and distribution of this document is restricted to the persons permitted by the security classification. This document is subject to
copyright protection. This document is to be used only for the purposes for which it has been supplied and information contained in it must not be
disclosed or in any other way made available, in whole or in part, to persons other than those permitted under the security classification without
the prior written approval of the Association.

Disclaimer
The GSMA makes no representation, warranty or undertaking (express or implied) with respect to and does not accept any responsibility for, and
hereby disclaims liability for the accuracy or completeness or timeliness of the information contained in this document. The information contained
in this document may be subject to change without prior notice.

Compliance Notice
The information contain herein is in full compliance with the GSMA Antitrust Compliance Policy.

This Permanent Reference Document is classified by GSMA as an Industry Specification, as such it has been developed and is maintained by
GSMA in accordance with the provisions set out GSMA AA.35 - Procedures for Industry Specifications.

V0.1 Page 1 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Table of Contents
1 Introduction 6
1.1 Scope of document 6
1.2 General considerations 7
1.3 Definition of Terms 7
1.4 Document Cross-References 9
2 QoS Parameters 11
2.1 QoS Parameters Model 11
2.2 Service-independent QoS parameters 12
2.2.1 Network Availability [2] [12] 14
2.2.2 Network Accessibility [2] 15
2.2.3 Circuit Switched LU Success Ratio (CS LU – SR) [2] 15
2.2.4 Circuit Switched Location Update Delay (CS LU - D) 16
2.2.5 Packet Switched LU Success Ratio (PS LU - SR) [2] [7] 17
2.3 Packet Switched Location Update Delay (PS LU - D) [2] 17
2.3.1 PDP Context Activation Success Ratio (SA PSD) [2] [19] 18
2.3.2 PDP Context Activation Time [2] [19] 19
2.3.3 PDP Context Cut-Off Ratio [2] [8] 19
2.3.4 Default EPS Bearer Context Activation Success Ratio [2] [3] 20
2.3.5 Default EPS Bearer Context Activation Time [2] 21
2.3.6 Dedicated EPS Bearer Context Activation Success Ratio [2] 22
2.3.7 Dedicated EPS Bearer Context Activation Time [2] 22
2.3.8 DNS Host Name Resolution Success Ratio [%] [2] 24
2.3.9 DNS Host Name Resolution Time [s] [2] 24
2.4 Telephony Service 25
2.4.1 Service Accessibility Telephony - MO (SA-T-MO) [2] [12] [14] 25
2.4.2 Service Accessibility Telephony - MT (SA-T-MT) [2] [14] 26
2.4.3 27
2.4.4 Setup Time Telephony – MO (ST-T-MO) [1] [12] [13] 27
2.4.5 Setup Time Telephony – MT (ST-T-MT) [2] [12] [13] 28
2.4.6 Abstract definition 28
2.4.7 CSSR - MO (Call Setup Success Ratio) [7] 29
2.4.8 Abstract definition 29
2.4.9 CSSR - MT (Call Setup Success Ratio) [7] 30
2.4.10 Abstract definition 30
2.4.11 REL (ISUPv2 signalling transparency) [16] 30
2.4.12 Abstract definition 30
2.4.13 Call Completion Ratio Circuit Switched Telephony (CCR-CS-T) [2] 31
2.4.14 ALOC (Average Length of a Call) [4] 32
2.4.15 CLI Transparency [5] 32
2.4.16 Speech Quality on Call Basis (SpQ) [2] [21] [27] [28] [29] 32
2.4.17 Abstract definition 32
2.4.18 CSFB Return to LTE Success Ratio 33
2.4.19 Trigger points 33

V0.1 Page 2 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.4.20 CSFB Return to LTE Time 34

2.4.21 Abstract definition 34
2.5 Short Message Service 34
2.5.1 Service Accessibility SMS MO (SA SMS MO) [2] 34
2.5.2 Service Accessibility SMS MT (SA SMS MT) [2] 35
2.5.3 Abstract definition 35
2.5.4 Access Delay SMS MO (AD SMS-MO) [2] 35
2.5.5 Abstract definition 35
2.5.6 Access Delay SMS MT (AD SMS-MT) [2] 36
2.5.7 Abstract definition 36
2.5.8 End-to-End Delivery Time for SMS MO (E2E DT SMS-MO) [2] 36
2.5.9 Abstract definition 36
2.5.10 Completion Ratio SMS (CR SMS) [2] [15] 37
2.5.11 Abstract definition 37
2.6 Circuit Switched Data Service 38
2.6.1 Service Accessibility, Circuit Switched Data (SA –CSD) [15] 38
2.6.2 Set-up Time (ST – CSD) 39
2.6.3 Abstract definition 39
2.6.4 Data Quality (DQ-CSD) 39
2.7 Packet Switched Data Service (General Packet Radio Service) 40
2.7.1 Service Accessibility Ratio – Packet Switched Data (SA – PSD) [2] [7] 40
2.7.2 Set-up Time – Packet Switched Data (ST – PSD) [2] 41
2.7.3 Abstract definition 41
2.7.4 Service Integrity - Throughput (Kbit/sec) [2] 42
2.7.5 Service Integrity - Goodput (Kbit/sec) [10] 42
2.7.6 Abstract definition 42
2.7.7 Service Integrity - Roundtrip Time [11] 42
2.7.8 Abstract definition 42
2.7.9 Service Integrity – Packet Loss [6] [11] 43
2.7.10 Completed Session Ratio (CoSeR – PSD) 43
2.7.11 Abstract definition 43
2.7.12 Service Retainability – Average PDP Context Average Session Time [8] 44
2.7.13 Abstract definition 44
2.8 Data Service Class Definitions and Measurements 44
2.8.1 Conversational Class Data 45
2.8.2 Abstract formula: 45
2.8.3 Trigger points: 45
2.8.4 Streaming Class 45
2.8.5 Abstract definition 45
2.8.6 Interactive Class 46
2.8.7 Abstract definition 46
2.8.8 Background class 47
2.8.9 Abstract definition 47
2.9 FTP QoS Parameters 47
2.9.1 FTP {Download|Upload} IP Service Access Success Ratio [%] [2] 47

V0.1 Page 3 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.9.2 FTP {Download|Upload} IP Service Access Setup Time [s] [2] 48

2.9.3 Abstract definition 48
2.9.4 FTP {Download|Upload} Session Success Ratio [%] [2] 49
2.9.5 Abstract definition 49
2.9.6 FTP {Download|Upload} Session Time [s] [2] 49
2.9.7 Abstract definition 49
2.9.8 FTP {Download|Upload} Mean Data Rate [kbit/s] [2] 50
2.9.9 Abstract Formula 50
2.9.10 FTP {Download|Upload} Data Transfer Success Ratio [%] [2] 50
2.9.11 Abstract definition 50
2.9.12 FTP {Download|Upload} Data Capacity 51
2.9.13 Abstract definition 51
2.10 WEB Browsing (HTTP / HTTPS) QoS Parameters 52
2.10.1 HTTP IP-Service Access Success Ratio [%] [2] 52
2.10.2 HTTP IP-Service Setup Time [s] [2] 52
2.10.3 Abstract definition 52
2.10.4 HTTP Session Success Ratio [%] [2] 53
2.10.5 Abstract definition 53
2.10.6 HTTP Session Time [s] [2] 53
2.10.7 Abstract definition 53
2.10.8 HTTP Mean Data Rate [kbit/s] [2] 53
2.10.9 Abstract definition 53
2.10.10 HTTP Data Transfer Success Ratio [%] [2] 54
2.10.11 Abstract definition 54
2.10.12 HTTP Content Compression Ratio [%] [2] 54
2.10.13 Abstract definition 54
2.10.14 HTTP {Download|Upload} Data Capacity 55
2.10.15 Abstract definition 55
2.11 PING QoS Parameters 55
2.11.1 PING Packet Loss Ratio 55
2.11.2 PING Round Trip Time [ms] [2] 56
2.11.3 Abstract definition 56
3 QoS Parameters for Interconnection 56
3.1 ASR = (Answer Seizure Ratio) 56
3.2 ABR = (Answer BID Ratio) 57
3.3 CLI = (Calling Line Identification) 57
3.4 NER = (Network Efficiency Ratio) 57
3.5 PGAD = (Post Gateway Delay) 57
3.6 PDD = (Post Dial Delay) 57
3.7 ALOC = (Average Length of Conversation) 57
3.8 Speech Quality on Sample Basis. 57
4 Reference: ITU-T Recommendation P.863 [27] VoLTE / ViLTE Quality
Parameters 57
4.1 VoLTE / ViLTE Service Quality Parameters 58
4.1.1 VoLTE / ViLTE Accessibility Parameters 58

V0.1 Page 4 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

4.1.2 VoLTE / ViLTE Service Quality Parameters 59

4.1.3 SRVCC (PS-CS) Quality Parameters [34] 66
4.2 VoLTE / ViLTE Networks & Transport QoS Parameters 66
4.2.1 VoLTE / ViLTE EPS Bearer Quality Parameters 66
4.2.2 VoLTE / ViLTE Media Transport Quality Parameters 69
4.2.3 DIAMETER Quality Parameters 73
5 5G NSA (Option 3.X) – Dual Connectivity 76
5.1 5G NSA Coverage Indication 76
Abstract definition 76
Abstract formula 76
5.2 NR RSRP 77
Abstract formula 77
5.3 DC NR Restricted 77
Abstract definition 77
Abstract formula 77
5.4 5G ENDC_Used allocation 77
Abstract definition 78
Abstract formula 78
5.5 HTTP Data Access Success ratio 78
5.6 MBB Download Rate Mbits/s 78
5.7 MBB Upload Rate Mbits/s 78
5.8 MBB Latency milliseconds 78
6 5G SA (Option 2) 79
6.1 VoNR – Voice Services 79
6.1.1 5G Signal to Noise 79
6.1.2 VoNR Call Quality Score 79
6.1.3 VoNR Call Setup Time 79
6.1.4 VoNR Call Drop Rate 79
6.1.5 Jitter 80
6.1.6 Packet Loss 80
6.1.7 Latency 80
6.1.8 Data Rate 81
6.1.9 5QI 81
6.1.10 SST 81
6.2 SMS over 5G 82
6.2.1 SMS Submission Success Rate 82
6.2.2 SMS Submission Delay 82
6.2.3 SMS End To End Delivery Success Rate 82
6.2.4 SMS End To End Latency 83
7 Single Service Indicator and Single Quality Indicator 83
7.1 Single Service Indicator 83
7.1.1 Purpose of SSI 83
7.1.2 SSI Calculation 83
7.2 Single Quality Indicator 85

V0.1 Page 5 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

7.2.1 SQI calculation 85

Annex A Examples for measuring trigger points 85
A.1 SMS-Service: 85
A.1.1 Layer 3 Messages: 85
Annex B Document Management 87
B.1 Document History 87
B.2 Other Information 88

1 Introduction

1.1 Scope of document

This document defines quality of service (QoS) parameters and their computation. A QoS
parameter is also called quality Key Performance Indicator (KPI). Although the QoS
definitions themselves are generic the scope of this document underlines the roaming
deployment of those QoS parameters.

The parameter definition is split into two parts: the abstract definition and the generic
description of the measurement method with the respective trigger points.

Consistent use of the definitions in this document will allow independent parties to compare
QoS measurements and results. Figure 1 shows the relationship to the other QoS-related
PRDs.

Figure 1: Relation of IR.42 to other PRDs in GRQ framework

V0.1 Page 6 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Figure 1 depicts the GRQ framework and the relationship of IR.42 with the other relevant
GSMA PRDs. The GSMA GRQ framework has been designed to standardize the
parameters, measurement and implementation of roaming quality KPIs and to provide the
guidelines and supporting information for operators to deploy those quality KPIs within end
to end roaming SLAs.

The framework consists of four PRDs and a trouble report database.

• IR.42 (the present PRD) contains the quality KPI definitions and calculations relevant
for the global roaming
• IR.81 specifies the roaming quality KPI test methods, test implementations and the
KPI thresholds for roaming
• BA.51 provides the specific guidelines and advanced agreements to those operators
wishing to establish an end-to-end Roaming Service Level Agreement (RSLA) either
between themselves and/or through a Roaming Hub Provider.
• AA.13 Annex C12 provides the basis for establishing Roaming Service Level
Agreement between operators.
• Operator Trouble Report – operator report to communicate issues on their network to
their roaming partners. May use the global NSIA (Network Service Interruption
Advisory) reporting via the GSMA Infocentre when available.

1.2 General considerations

All the defined QoS parameters and their computations are based on field measurements.
That indicates that the measurements were made from user’s point of view (full end-to-end
perspective, taking into account the needs of testing).

It is assumed that the mobile and the desired service can be operated correctly by the end
user, as operability is not evaluated in this document. For the purpose of measurement it is
assumed that:

• the service is available and not barred for any reason;

• routing is defined correctly without errors and;
• the target subscriber equipment is ready to answer the call.
Only voice quality values measured for calls ended successfully must be used for statistical
analysis. However, measured values from calls ended unsuccessfully (for example, calls that
are dropped) should be available for additional evaluation if required, and therefore must be
stored. Further preconditions will apply when reasonable.

Monitoring of services using the parameters defined in this document could impact the traffic
and load on the networks involved, including the home public mobile network (HPMN), the
visited public mobile network (VPMN), and intermediary networks. Due consideration must
therefore be given to the monitoring regime to avoid unnecessary or adverse impacts on
these networks.

1.3 Definition of Terms

Term Meaning
5GC 5G Core Network
AMF Access & Mobility Function

V0.1 Page 7 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Term Meaning
APN Access Point Name
AUSF Authentication Server Function
CS Circuit Switched
CSFB Circuit Switched Fall Back
DNS Domain Name System
ENDC Enabling Dual Connectivity
EPC Evolved Packet Core
FTP File Transfer Protocol
GBR Guaranteed Bit Rate
gNB g-node 5G radio access station
HLR Home Location Register
HPMN Home Public Mobile Network
HSS Home Subscriber Server
HTTP Hypertext Transport Protocol
IMS Internet Protocol Multimedia Subsystem
IPX Internet Protocol Exchange
ISUP ISDN User Part
KPI Key Performance Indicator
LQO Listening Quality Objective
LTE Long Term Evolution (Radio)
MME Mobility Management Entity
MT Mobile Terminated
NER Network Efficiency Ratio
MS Mobile Station
MSC Mobile Switching Centre
NFV Network Function Virtualisation
NR New Radio (provided by gNB)
OCN Original Called Number
PCEF Policy & Charging Enforcement Function
PCF Policy Control Function
PCRF Policy & Charging Rule Function
P-CSCF Proxy - Call Session Control Function
PESQ Perceptual Evaluation of Speech Quality
PEVQ Perceptual Evaluation of Video Quality
PING Packet Internet Groper

POLQA Perceptual Objective Listening Quality Analysis

PRD Permanent Reference Document

V0.1 Page 8 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Term Meaning
QCI QoS Class Indicator
QoS Quality of Service
RDN Redirecting Number
RTCP RTP Control Protocol
RTP Real-time Transport Protocol
RSRP Reference Signal Received Power
RSRQ Reference Signal Received Quality
SGSN Serving GPRS Support Node
SIP Session Initiation Protocol
SLA Service Level Agreement
SMF Session Management Function
SMS MO Mobile Originated SMS
SMS MT Mobile Terminated SMS
SMSoSGs SMS over SGs
SS7 Signalling System 7
SQI Service Quality Indicator
SRVCC Single Radio Voice Call Continuity
SSI Single Service Indicator
UDM Unified data Function
UPF User Plane Function
ViLTE (conversational) Video over LTE
VoLTE Voice over Long Term Evolution
VPMN Visited Public Mobile Network

1.4 Document Cross-References

The following documents contain provisions which, through references in this text, constitute
provisions of the present document.

• References are non-specific, i.e. refer to the latest version of the document.
No. Document Description
[1] ETSI TS 102 250-3 "Speech Processing, Transmission and Quality Aspects (STQ);
QoS aspects for popular services in GSM and 3G networks; Part 3:
Typical procedures for Quality of Service measurement equipment".
[2] ETSI TS 102 250-2 "Speech Processing, Transmission and Quality Aspects (STQ);
QoS aspects for popular services in GSM and 3G networks; Part 2:
Definition of Quality of Service parameters and their computation ".
[3] GSMA PRD IR.88 LTE and EPC Roaming Guidelines
[4] ITU-T-E.437 “Comparative metrics for network performance Management”
[5] ETSI EN 300 089 "Integrated Services Digital Network (ISDN); Calling Line
Identification Presentation (CLIP) supplementary service; Service
description".

V0.1 Page 9 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

No. Document Description

[6] ITU-T-Y.1540 “IP packet transfer and availability performance parameters”; IPLR
– IP Packet Loss Ratio
[7] 3GPP TS 29.002 ”Mobile Application Part (MAP) specification”
[8] 3GPP TS 29.060 ”General Packet Radio Service (GPRS); GPRS Tunnelling Protocol
(GTP) across the Gn and Gp interface”
[9] 3GPP TS 24 008 "Mobile radio interface Layer 3 specification; Core network
protocols; Stage 3".
[10] IETF RFC 2647 ”Benchmarking Terminology for Firewall Performance; 3.1.7
Goodput”
[11] GSMA PRD IR.34 GSMA Inter-Service Provider IP Backbone Guidelines
[12] ITU-T-E.800 “Terms and Definitions Related to Quality of Service and Network
Performance Including Dependability - Telephone Network and
ISDN Quality of Service, Network Management and Traffic
Engineering (Study Group II)”
[13] ITU-T E.431 “Service Quality Assessment for Connection Set-up and Release
Delays (Study Group II)”
[14] ITU-T E.425 “Internal Automatic Observations Series E: Overall Network
Operation, Telephone Service, Service Operation and Human
Factors Network Management - Checking the Quality of the
International Phone Service”
[15] GSMA PRD BA.51 “Roaming Service Level Agreement Guidelines”
[16] ITU-T QE.850 “Usage of cause and location in the digital subscriber signalling
system No. 1 and the signalling system No. 7 ISDN user part”
[17] ITU-T Q.732.2 “Stage 3 Description for Call Offering Supplementary Services
Using Signalling System No.7: Call Diversion Services: - Call
Forwarding Busy - Call Forwarding No Reply - Call Forwarding
Unconditional - Call Deflection - Series Q: Switching and Signalling
- Specifications of Signalling System No.7 - ISDN Supplementary
Services”
[18] GSMA PRD IR.25 VoLTE Roaming Testing
[19] GSMA PRD IN.21 “GSM Association Roaming Database, Structure and Updating
Procedures”
[20] 3GPP TS 29.272 MME and SGSN related interfaces based on Diameter protocol
[21] ITU-T P.862 Perceptual evaluation of speech quality (PESQ): An objective
method for end-to-end speech quality assessment of narrow-band
telephone networks and speech codecs
[22] GSMA PRD AA.13 “International Roaming Agreements – Common Annexes
[24] 3GPP TS 45.008 Radio subsystem link control
[25] 3GPP TS 25.304 UE procedures in idle mode and procedures for cell reselection in
connected mode
[26] 3GPP TS 36.304 EUTRA; UE procedures in idle mode
[27] ITU-T P.863 Perceptual Objective Listening Quality Assessment (POLQA)
[28] ETSI TR 102 506 Estimating Speech Quality per Call
[29] ITU-T P.862.1 Perceptual evaluation of speech quality (PESQ): Mapping function
for transforming P.862 raw result scores to MOS-LQO

V0.1 Page 10 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

No. Document Description

[30] GSMA PRD IR.65 IMS Roaming & Interworking Guidelines
[31] GSMA PRD IR.92 IMS Profile for Voice and SMS
[32] GSMA PRD IR.94 IMS Profile for Conversational Video Service
[33] GSMA PRD VoLTE-RCS Roaming and Interconnection Guidelines
NG.103
[34] ETSI TR 103 219 QoS Aspect of Voice Communication in an LTE Environment
[35] ITU-T G.107.1 Wideband E-Model
[36] 3GPP TS 29.109 Generic Authentication Architecture (GAA); Zh and Zn Interfaces
based on Diameter Protocol
[37] 3GPP TS 29.213 Policy and Charging Control Signalling Flows and Quality of Service
(QoS) Parameter Mapping
[38] 3GPP TS 29.214 Policy and Charging Control over Rx Reference Point
[39] 3GPP TS 29.229 Cx and Dx interfaces based on the Diameter protocol; Protocol
Details
[40] 3GPP TS 29.329 Sh interfaces based on the Diameter protocol; Protocol Details
[41] ITU-T J.247 Objective perceptual multimedia video quality measurement in the
presence of a full reference
[42] IETF RFC 3550 RTP: A Transport Protocol for Real-Time Applications
[43] IETF RFC 4961 Symmetric RTP / RTP Control Protocol (RTCP)
[44] ITU-T Y.1540 Internet protocol data communication service – IP packet transfer
and availability performance parameters
[45] IETF RFC 5481 Packet Delay Variation Applicability Statement
[46] 3GPP TS 24.301 NAS Protocol for Evolved Packet System

2 QoS Parameters

2.1 QoS Parameters Model

Figure 2 shows a model for QoS parameters. This model has four layers.

• The first layer is the Network Availability, which defines QoS from the viewpoint of the
network and service provider.
• The second layer is the Network Access, the basic requirement for all the other QoS
aspects and QoS parameters. The outcome of this layer is the QoS parameter
Network Accessibility from the viewpoint of the service user.
• The third layer contains the three QoS aspects: Service Access, Service Integrity and
Service Retain-ability.
• The fourth layer contains the different services to be provided to the service user in
case of roaming, for example Voice (Telephony), SMS, Web browsing, File
Transferring and Ping services. The outcomes are the QoS parameters for those
services.
For monitoring of roaming QoS, please refer to PRD IR.81 for details of call flows and trigger
points.

V0.1 Page 11 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Note that the QoS is dependent on interconnectivity between operators and carriers or
between roaming hub providers. An Internet Protocol Exchange (IPX) QoS monitoring
scheme provides more possibilities for measuring QoS across IPX interconnections [11].

Figure 2: Four Layer model for QoS Parameters.

2.2 Service-independent QoS parameters

The service-independent QoS parameters characterise network availability, network
accessibility, IP connectivity and bearer context cut-off. The parameters are therefore
considered as technology-dependent. Table 1 provides an overview.

V0.1 Page 12 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Technologies
2G / 3G 4G VoLTE/ViLTE 5G NSA Future
&5GSA technology
Network Scan for Scan for EPS Network FFS
/service PLMN: PLMN: feature support: Scan for
availability Radio Radio IMS voice / PLMN: 5G
network network video over PS Radio
availability availability session in S1 network
mode presence
supported
Network / CS and/or PS LTE attach IMS NSA Dual
service attach: (EPS or Registration Connectivity
accessibility Network EPS+IMSI): success ratio allocation
selection and Network /time (ENDC)
registration selection and
successful registration
ratio /time successful
ratio /time
IP connectivity (Primary) Default / QCI 5 default Default /
PDP context Dedicated EPS bearer Dedicated
activation: EPS bearer context EPS bearer
PDP context context activation for context
activation activation: the IMS well- activation:
successful Default EPS known APN Default EPS
ratio /time bearer connection; bearer
context QCI 1 context
activation dedicated EPS activation
successful bearer context successful
ratio /time activation for ratio /time
voice media;
Default /
dedicated EPS
bearer context
activation
success ratio
/time
Network / PDP context Default EPS QCI 5 default / Default EPS
service dropped: bearer QCI 1 bearer
retainability PDP context context dedicated EPS context
cut-off ratio dropped: bearer context dropped:
Default EPS cut-off ratio Default EPS
bearer bearer
context cut-off context cut-
ratio off ratio
Table 1: Overview of QoS parameters on technology-dependency

For 2G/ 3G (i.e. GSM/GPRS and UMTS) networks it is necessary to establish a data
connection before the possibility of accessing a service. In LTE networks the concept of IP
connectivity “always on” has been established. It allows a faster access to the services of a
mobile network by state changes on request of a user or application. The impact of this
concept on the QoS parameters is that they cannot be triggered explicitly anymore as they

V0.1 Page 13 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

require a specific service request. Nevertheless the defined QoS parameters remain valid
and can be derived from the trigger points given.

This section contains also DNS KPI. DNS service is an intermediate internet service used by
the other specific PS data services such as FTP or HTTP. However, the DNS service is
independent from those specific data services. Network Accessibility Circuit Switched (NA –
CS) [1] [12].

In 5G NSA (option 3.X) the network needs to establish a LTE attach and validates the Dual
Connectivity restriction for the subscriber profile, then it will handle the hand-shake for
establish the corresponding bearer with enabling dual connectivity.

Please refer to GSMA PRD IR.81 for details of call flow and trigger points.

2.2.1 Network Availability [2] [12]

Abstract definition
Probability that the Mobile Services are offered to a user.

See ITU-T Rec. E.800: The probability that the user of a service after a request receives the
proceed-to-select signal within specified conditions.

See complementary QoS parameter RNU in ETSI TS 102 250-2 (5.1 Radio Network
Unavailability).

Computation

Trigger points:
GSM: C1-Criteria > 0 [24]

GPRS: broadcasting GPRS indicator in system information 3 or system information 4

UMTS (WCDMA) [25], LTE [26]: S Criteria fulfilled

Any emergency camping on any other than the target networks is considered as no network.

The target networks could constitute more than one network, for example to cover national
or international roaming or via a roaming hub.

When the mobile equipment supports multi-mode (GSM/UMTS/LTE/5G),the judgement on

Radio Network Availability is made with respect to the radio access technology under the
test aspect.

For CSFB and SMSoSGs, LTE S criteria is applied.

For 5G NSA which relies on LTE network to establish the control plane. The criteria will
involve the dual connectivity enable by network (ENDC Available), which will mean the 5G
NR is being broadcasted and detected by UE as 5G indicator. (see section 5.1)

For 5G SA, the availability of Radio Technology (RAT) will be done from UE network scan in
similar way than others legacy technology following previous statement (!*!)

V0.1 Page 14 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula:

probing attempts with mobile services available

Radio Network Availability [%] =  100
all probing attempts

2.2.2 Network Accessibility [2]

Abstract definition
Probability that the user performs a successful selection and registration on the desired
PLMN (manual selection mode, automatic selection mode with a defined desired PLMN) or
on a certain PLMN (automatic selection mode without a defined desired PLMN).

See ETSI 102 250-2 (5.2.1 Network Selection and Registration Failure Ratio).

Computation

Trigger points:
Initiate manually or automatically PLMN selection, stop measurement after successful
registration.

Abstract formula:

NA[%] = {Manual | Automatic} Network Selection and Registration Success Ratio [%] =
successful selection and registration attemps on PLMN
 100
all selection and registration attempts

NA-CS[%] is applied to GSM or UMTS-CS. NA-PS[%] is applied to GPRS, UMTS-PS, LTE,

respectively.

2.2.3 Circuit Switched LU Success Ratio (CS LU – SR) [2]

The CSLU success ratio describes the probability that a subscriber can successfully attach
to the CS network.

See ETSI 102 250-1 (5.2 Network non-accessibility).

Computation
See ETSI TS 102 250-2 (5.2 Network non-accessibility).

Abstract formula:

NrSuccessfulCSAttachAttempts
CSLUSuccessRatio = x100%
TotalNrCSAttachAttempts

V0.1 Page 15 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Trigger points:
Start: Mobile sends the CS attach request message.

Stop: Mobile receives the CS attach accept message.

Remarks
Success ratio measurements will depend on whether the LU is the very first LU attempt or
one of subsequent attempts. (See remarks for CSLU-D below).

2.2.4 Circuit Switched Location Update Delay (CS LU - D)

Abstract definition
This CSLU delay describes the time period needed to attach to the CS network.

Computation

Abstract formula:

Sum(t CSAttachComplete − t CSAttachStart )

CSLUDelay =
NrSuccessfulCSAttachAttempts

Trigger points:
(for the computation of the unit CSLU delay):

Start: Point of time when the mobile sends the attach

request message.
Stop: Point of time when the mobile receives the attach
accept message.

Remarks:
The difference between an attach of a known subscriber and an unknown subscriber will be
reflected in the time period indicating the attach setup time. In case of an unknown
subscriber (meaning that the Mobile Switching Centre (MSC) has changed since the detach,
or if it is the very first attach of the mobile to the network), the MSC contacts the Home
Location Register (HLR) in order to receive the subscriber data. The attach setup time of an
unknown subscriber will be slightly longer than the one of a known subscriber.

While determining the average attach setup time only successful attach attempts are
included in the calculations.

V0.1 Page 16 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.2.5 Packet Switched LU Success Ratio (PS LU - SR)1 [2] [7]

Abstract definition
The PSLU success ratio describes the probability of a subscriber to successfully attach to
the PS network.

See ETSI TS 102 250-2 (5.3 Attach Failure Ratio)

Computation 4

Abstract formula:

NrSuccessfulPSAttachAttempts
PSLUSuccessRatio = x100%
TotalNrPSAttachAttempts

Trigger points:
Start: Mobile sends the PS attach request message

Stop: Mobile receives the PS attach accept message.

Remarks
Depending upon the technologies, PS attach is understood as either a GPRS attach, or an
EPS attach.

A combined EPS + IMSI attach is applied to CSFB or SMSoSGs.

2.3 Packet Switched Location Update Delay (PS LU - D) [2]

Abstract definition
This PSLU delay describes the time period needed to attach to the PS network.

See ETSI TS 102 250-2 (5.4 Attach Setup Time)

Computation

Abstract formula:

Sum(t PSAttachComplete − t PSAttachStart )

PSLUDelay =
NrSuccessfulPSAttachAttempts

1The KPI Packet-Switched Location Update was named originally from the
MAP_Update_GPRS_Location procedure, as described in 3GPP TS 29.002 [7]. It is used for the PS
attach, i.e. for GPRS attach. The same term Location is also used in the Location Management
procedure in 3GPP 29.272 [20] for Update Location and Cancel Location at S6a, i.e. EPS attach.

V0.1 Page 17 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Trigger points:
Start: Point of time when the mobile sends the attach
request message

Stop: Point of time when the mobile receives the attach

accept message.

Remarks:
The difference between an attach of a known subscriber and an unknown subscriber will be
reflected in the time period indicating the attach setup time. In case of an unknown
subscriber (meaning that the SGSN, S4 SGSN or MME has been changed since the detach,
or if it is the very first attach of the mobile to the network), the SGSN contacts the HLR, the
S4 SGSN contacts HSS or the MME contacts HSS respectively, in order to receive the
subscriber data. The attach setup time of an unknown subscriber will be slightly longer than
the one of a known subscriber.

2.3.1 PDP Context Activation Success Ratio (SA PSD) [2] [19]

Abstract definition
A packet-switch data session will be considered set-up successfully if a PDP Context can be
successfully activated.

See ETSI TS 102 250-2 (5.5 PDP Context Activation Failure Ratio)

Computation

Abstract formula:

NrOfPDPContextActivationsSuccessful
PDPContext ActivationSuccessRate = 100% 
NrOfAllPDPContextActivationAttempts

Any PDP Context Activation request is considered as an attempt.

Trigger points:
PDP Context Activations are considered successful upon the reception of notifications of
successful PDP context activation (Activate PDP Context Accept).

Remarks:
It is recommended to use a limited and defined list of APNs for consistent measurements.
Operators are encouraged to maintain a list of standard APNs for measurements in PRD
IR.21 (or alternatively defined in roaming service level agreements (SLAs)). The list could
include APNs for MMS, WAP and/or internet, both pre-paid and post-paid. The APNs should
be defined as part of test conditions.

LTE: see Default EPS Bearer Context Activation Success Ratio

V0.1 Page 18 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.3.2 PDP Context Activation Time [2] [19]

Abstract definition
Is time between sending the PDP Context Activation request and receiving the notification of
successful completion of that activation.

See ETSI TS 102 250-2 (5.6 PDP Context Activation Time)

Computation

Abstract formula:

( )
PDP Context Activation Time [s] = t PDP context activation accept - t PDP context activation request [s]

Trigger points:
Start: Sending of the PDP Context Activation request

End: Reception of the notification of successful PDP

context activation (Activate PDP Context Accept)

Remarks: 5
When averaging the PDP Context Activation Time, only the successful activations should be
considered.

It is recommended to use a limited and defined list of APNs for consistent measurements.
Operators are encouraged to maintain a list of standard APNs for measurements in PRD
IR.21 (or alternatively defined in Roaming SLAs). The list could include APNs for MMS, WAP
and/or internet, both pre-paid and post-paid. The APNs should be defined as part of test
conditions.

User activation (GPRS PDP context creation procedure) is recommended as network

activation tends to be for local services (home usage).

LTE: See Default EPS Bearer Context Activation Time or additional Default EPS Bearer
Context Activation Time

2.3.3 PDP Context Cut-Off Ratio [2] [8]

Abstract definition
The PDP context cut-off ratio denotes the probability that a PDP context is deactivated
without being deactivated intentionally by the user.

ETSI TS 102 250-2 (5.7 PDP Context Cut-off Ratio).

A precondition for measuring this parameter is that a PDP context is successfully

established first.

2.3.4 Default EPS Bearer Context Activation Success Ratio [2] [3]

Abstract definition
Default EPS bearer context for the default APN: success of the EPS attach procedure (PS
LU-SR)
In case of an additional default EPS bearer context: success of activation of the additional
default EPS bearer context

See ETSI TS 102 250-2 (5.12.1 Default EPS Bearer Context Activation Failure Ratio)

Computation

Abstract formula:
Default EPS bearer context activation success ratio for the default APN: same as EPS
attach success ratio (PS LU-SR).

Additional default EPS bearer context, referring to an APN different from the default one:

Additional Default EPS Bearer Context Activation Success Ratio [%] =

additional PDN connection establishment successes
 100
additional PDN connection initiations

V0.1 Page 20 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Any additional Default Context Activation request is considered as an initiation.

Trigger points:
The trigger points for the default EPS bearer context referring to the default APN is same as
EPS attach procedure.

PDN Connectivity Request for additional default EPS bearer Context Activations is sent by the
UE. The additional PDN connectivity is considered successful upon the reception of
notifications of additional default EPS bearer context activation (Activate default EPS bearer
Context Request)

When the UE is powered on it will perform an EPS attach which includes the registration with
the network and the setup of a default EPS bearer context for the default APN. A default
EPS bearer context is successfully established when (EPS) ATTACH COMPLETE is sent by
the UE. This context replaces the primary PDP context defined for 2G/3G networks.

2.3.5 Default EPS Bearer Context Activation Time [2]

Abstract definition
The default EPS bearer context activation time is the time period needed to establish the
initial default EPS bearer context for the default APN or any additional PDN connection (i.e.
any additional default EPS bearer context), respectively.

See ETSI TS 102 250-2 (5.12.2 Default EPS Bearer Context Activation Time)

Computation

Abstract formula:
Default EPS bearer context for the default APN2:

Default EPS Bearer Context Activation Time [s] = (t attach complete - t attach request )[s]

Additional default EPS bearer context:

2The PDN CONNECTIVITY REQUEST message is piggybacked in ATTACH REQUEST. The ACTIVATE
DEFAULT EPS BEARER CONTEXT REQUEST message is piggybacked in ATTACH ACCEPT. The ACTIVATE
DEFAULT EPS BEARER CONTEXT ACCEPT message is piggybacked in ATTACH COMPLETE.

V0.1 Page 21 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Additional Default EPS Bearer Context Activation Time [s] =

(t Additional PDN connection establishm ent - t Additional PDN connection initiation ) [s]

Trigger points:
PDN CONNECTIVITY REQUEST for additional default EPS bearer Context Activations is
sent by the UE. The ACTIVATE DEFAULT EPS BEARER CONTEXT ACCEPT is sent by
the UE.

Remarks:
Precondition: Successful default PDN connection for the default APN is prior to the
additional default PDN connections.

2.3.6 Dedicated EPS Bearer Context Activation Success Ratio [2]

Abstract definition
The dedicated EPS bearer context activation success ratio measures the probability that a
dedicated bearer can be activated. It is the proportion of successful dedicated bearer context
activation attempts and the total number of dedicated bearer activation attempts.

See ETSI TS 102 250-2 (5.13.1 Dedicated EPS Bearer Context Activation Failure Ratio)

Computation

Abstract formula:

Dedicated EPS Bearer Context Activation success Ratio [%] =

dedicated EPS bearer activation successes
 100
dedicated EPS bearer activation initiations

Trigger points:
The UE requests bearer resource allocation (or modification) procedure (BEARER
RESOURCE ALLOCATION REQUEST or BEARER RESOURCE MODIFICATION
REQUEST). The activation of the dedicated EPS bearer is considered as initiated when the
ACTIVATE DEDICATED EPS BEARER CONTEXT REQUEST is received by the UE.

Remarks:
Precondition: Successful PDN connection for the APN, for which a Dedicated EPS Bearer
shall be established, already exists

2.3.7 Dedicated EPS Bearer Context Activation Time [2]

Abstract definition
The Dedicated EPS bearer context activation time is the time that is needed to establish a
dedicated bearer for user data transfer.

V0.1 Page 22 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

See ETSI TS 102 250-2 (5.13.2 Dedicated EPS Bearer Context Activation Time)

Abstract formula:

Dedicated EPS Bearer Context Activation Time [s] =

(t Dedicated EPS Bearer context activation success - t Dedicated EPS Bearer context activation initiation ) [s]

Trigger points:
BEARER RESOURCE ALLOCATION (or MODIFICATION) REQUEST is sent by the UE.
The ACTIVATE DEDICATED EPS BEARER CONTEXT ACCEPT is sent by the UE.

Remarks:
The dedicated EPS bearer context activation procedure is triggered by the UE for bearer
resource allocation and completed when the allocated dedicated EPS bearer context is
accepted by the UE.

EPS Bearer Context Cut-off Ratio [%] [2]

Abstract Definition
The default or dedicated EPS bearer context cut-off ratio measures whether a default or a
dedicated EPS bearer context is deactivated without being initiated intentionally by the user3.

See ETSI TS 102 250-2 (5.12.3 Default EPS Bearer Context Cut-off ratio, 5.13.3 Dedicated
EPS Bearer Context Cut-off ratio)

Computation

Abstract Formula

Default | Dedicated EPS Bearer Context Cut - off Ratio [%] =

default | dedicated EPS bearer context losses not initiated by the user
 100
successfully activated default | dedicated EPS bearer contexts

Trigger Points
Different trigger points for a EPS bearer context deactivation not initiated intentionally by the
user are possible. The EPS bearer context will be deactivated by the network. The UE
receives from the network the message DEACTIVATE EPS BEARER CONTEXT
REQUEST, MODIFY EPS BEARER CONTEXT REQUEST or DETACH REQUEST.

Remarks
Default EPS bearer context (for the default APN) loss means UE EPS detach.

3The default EPS bearer context for the default APN is indispensable to be maintained to use any
service over LTE networks or any additional default EPS bearer context. Deactivating the last default
EPS bearer will cause the UE detach.

V0.1 Page 23 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.3.8 DNS Host Name Resolution Success Ratio [%] [2]

Abstract Definition
The DNS host name resolution success ratio is the probability for a host name to host
address translation of a DNS resolver is successful.

See ETSI TS 102 250-2 (5.10 DNS Host Name Resolution Failure Ratio)

Computation

Abstract Formula

DNS Host Name Resolution Success Ratio [%] =

successful DNS host name resolution requests
 100
DNS host name resolution requests

Trigger points
Start: Request to resolve a host address from DNS server, or DNS protocol data packet
containing DNS type A (host address) “Standard query” query for the desired host name.

Stop: Host address received from DNS server, or DNS protocol data packet received
containing a type A (host address) “Standard query response, No error” response, the
respective type A “Standard query” query and an answer including the desired host name to
host address translation.

Remarks
Precondition for measurement:

The resolver shall not have direct access to any local DNS name server or any name
server’s zone.

Since messages carried by UDP are restricted to 512 bytes. UDP is the recommended
method for standard queries on the Internet.

The KPI is relevant only for PS services.

2.3.9 DNS Host Name Resolution Time [s] [2]

Abstract Definition
The DNS host name resolution time is the time it takes a host name to host address
translation.

See ETSI TS 102 250-2 (5.11 DNS Host Name Resolution Time)

Computation
The DNS host name resolution time is the time it takes a host name to host address
translation.

V0.1 Page 24 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract Formula

( )
DNS Host Name Resolution Time [s] = t StandardQu eryRespons e - tStandardQu ery [s]

Trigger points
Start: Request to resolve a host address from DNS server, or DNS protocol data packet
containing DNS type A (host address) “Standard query” query for the desired host name

Remarks
Precondition for measurement:

The resolver shall not have direct access to any local DNS name server or any name
server’s zone.

For static measurement methodologies, as defined in TS 102 250-3 [1], the queried DNS
name server shall have any data related to the host name to be resolved available as
authoritative data in one of the name server’s zones, so that no recursive lookups have to be
performed and no use of cached information will be required.

If the related data is not stored locally in the name server’s zone, the resolution time would
vary due to DNS caching strategies

The KPI is relevant only for PS services.

2.4 Telephony Service

To simplify the description of the voice call KPIs, R-party and H-party are used in the context
where R-party is the roaming side in VPMN and H-party is in HPMN.

A precondition for the applicability of this section to the voice CSFB:

The LTE networks (VPMN) support voice CSFB. The UE (represented by R-party) are
configured as voice centric and CS voice only or CS voice preferred, IMS PS Voice as
secondary.

2.4.1 Service Accessibility Telephony - MO (SA-T-MO) [2] [12] [14]

Abstract definition
Probability that the end-user can access the Mobile Telephony Service when requested if it
is offered by display of the network indicator on the Mobile Equipment.

See ITU-T Rec. E.800: The probability that a service can be obtained within specified
tolerances and other given operating conditions when requested by the user. The term NER
(Network Effectiveness Ratio defined in ITU-E 425) is understood as Service Accessibility
Telephony.

V0.1 Page 25 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

See complementary Service Non-Accessibility Telephony (SNAT) in ETSI TS 102 250-2

(6.6.1 Service non-accessibility).

Computation
For a successful call attempt:

CS, CSFB
1. The user hears the alerting tone, or R-party receives ALERTING
2. H-party rings, or sends ALERTING

Abstract formula:
Number of successful call attempts
Service Accessibility Telephony [%] = *100%
Number of call attempts

Trigger points:
Beginning of the call Successful pressing send button (it is important to
attempt:: check, if coverage has been given when the send
button is pressed, otherwise this Call Attempt
counts to Network Non Accessibility (NNA)).

Successful call attempt: Measurement (alerting4 tone heard or ALERTING

(CC message) received by R-party), and

H-party rings or sends ALERTING (CC message).

Remarks:
SAT/NER is not catching the voice carriers that are sending fake ring tones like sending fake
ACM/CPG messages.

2.4.2 Service Accessibility Telephony - MT (SA-T-MT) [2] [14]

Probability that the end-user can access the Mobile Telephony Service when requested if it
is offered by display of the network indicator on the Mobile Equipment.

See complementary Service Non-Accessibility Telephony (SNAT) in ETSI TS 102 250-2

(6.6.1 Service non-accessibility).

4Due to network problems and despite H-party being not busy, it may even be possible for the
R-party to receive a busy or not reachable signal. In this case, since no ALERTING message will be
sent, the test sample will be treated as a failure.

V0.1 Page 26 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation
For a successful call attempt:
CS, CSFB
The user hears the ringing or R-party sends ALERTING (CC message)

H-party hears the alerting tone or receives ALERTING (CC message)

Abstract formula:
Number of successful call attempts
Service Accessibility Telephony [%] = *100%
Number of call attempts

Trigger points:
Beginning of the call Successful pressing send button (it is important to
attempt of H-party: check, if coverage has been given when the send
button is pressed, otherwise this Call Attempt
counts to Network Non Accessibility (NNA))

CS, CSFB

Successful call attempt: Measurement (R-party rings or sends ALERTING

(CC message)

Alerting tone heard or ALERTING (CC message)

received by H-party).

2.4.3

2.4.4 Setup Time Telephony – MO (ST-T-MO) [1] [12] [13]

Abstract definition
Time between sending of complete address information and receipt of call setup notification
as defined in ETSI TS 102 250-2 (6.6.2 Setup Time).

See ITU-T Rec. E.800: The expectation of the time duration between an initial bid by the
user for the acquisition of a service and the instant of time the user has access to the
service, the service being obtained within specified tolerances and other given operating
conditions.

The term ST-T can also be understood as PDD (Post Dialling Delay); see ITU-E 431.

Computation
For a successful call attempt:

V0.1 Page 27 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

CS, CSFB

• The user hears the alerting tone, or R-party receives ALERTING

• H-party rings or sends ALERTING (CC message).

Abstract formula:
Setup Time Telephony [s] = t 2 - t 1

t2: point of time where signalling connect is established (alerting tone is heard or ALERTING
(CC message) is received by test equipment))

t1: point of time where the user presses the send button on mobile equipment

Trigger points:
Beginning of Setup Time Successful pressing send button at R-party (it is
measurement: important to check, if coverage has been given,
otherwise this Call Attempt counts to Network
Non Accessibility (NNA))

CS, CSFB

Successful signalling Measurement (alerting tone heard or ALERTING

connection: (CC message) received by R-party).

2.4.5 Setup Time Telephony – MT (ST-T-MT) [2] [12] [13]Abstract definition

Time between sending of complete address information and receipt of call setup notification
as defined in ETSI TS 102 250-2 (6.6.2 Setup Time).

ITU-T Rec. E.800: The expectation of the time duration between an initial bid by the user for
the acquisition of a service and the instant of time the user has access to the service, the
service being obtained within specified tolerances and other given operating conditions.

The term ST-T can also be understood as PDD (Post Dialling Delay); see ITU-E 431.

Computation

Abstract formula:
Setup Time Telephony [s] = t 2 - t 1

t2: point of time where connect is established (for example alerting5 or subscriber busy is
detected by test equipment))

t1: point of time where the customer presses the send button on mobile equipment

5If an end to end connection is not established, this measurement must be ignored. It is assumed that
early traffic channel assignment is used.

V0.1 Page 28 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

t2: point of time where signalling connect is established (for example ringing6 or ALERTING
(CC message) sent by test equipment))

t1: point of time where the user presses the send button on mobile equipment

Trigger points:
Beginning of Setup Time Successful pressing send button at H-party (it is
measurement: important to check, if coverage has been given,
otherwise this Call Attempt counts to Network
Non Accessibility (NNA))

CS, CSFB

Successful signalling Measurement (ringing or ALERTING (CC

connection: message) sent by R-party).

2.4.7 CSSR - MO (Call Setup Success Ratio) [7]Abstract definition

CSSR expresses the relationship between the number of seizures and the sum of the
number of seizures resulting in a successful established call.

Call Setup Success Ratio is defined in 3GPP TS 29 002.

Computation

Abstract formula:
Number of successful call establishments
CSSR [%] = *100%
Number of call attempts

Trigger points:
Beginning of the call Successful pressing send button (it is important to
attempt of R-party: check, if coverage has been given when the send
button is pressed, otherwise this Call Attempt
counts to Network Non Accessibility (NNA)).

Successful call Open connection between R-party and H-party,

establishment: where both parties can hear each other.

CS, CSFB

R-party receives CONNECT and H-party receives

CONNECT ACKNOWLEGE (CC messages)

6If an end to end signalling connection is not established, this measurement must be ignored. It is
assumed that early traffic channel assignment for GSM is used.

V0.1 Page 29 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.4.9 CSSR - MT (Call Setup Success Ratio) [7]Abstract definition

CSSR expresses the relationship between the number of seizures and the sum of the
number of seizures resulting in a successful established call.

CSSR is defined in 3GPP 29 002.

Computation

Abstract formula:
It is assumed that the A Party is in the home network in order to avoid an international
tromboning of the call between the VPMN and the HPMN.

Number of successful call establishments

CSSR [%] = *100%
Number of call attempts

Successful call Open connection between R-party and H-party,

establishment: where both parties can hear each other.

CS, CSFB

R-party receives CONNECT ACKNOWLEGE and

H-party receives CONNECT (CC messages)

2.4.11 REL (ISUPv2 signalling transparency) [16]Abstract definition

Effective uncorrupted transmission by the VPMN of the Cause Value in the Release (REL)
ISDN User Part (ISUP) messages, as defined in ITU-T Q.850.

Testing protocol:
3. The HPMN sends to the VPMN a REL ISUP message with a valid populated ‘Cause
Value’ field
4. The VPMN must send back to the HPMN the REL ISUP message with the same
‘Cause Value’
5. The HPMN uses SS7 monitoring tool to measure the key performance indicator (KPI)

V0.1 Page 30 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation

Abstract formula:
Number of uncorrupted Cause Value in REL ISUP messages
REL− CV [%] =  100%
Number of sent REL ISUP messages with populated Cause Value

Effective uncorrupted transmission by the VPMN of Original Called Number, (OCN)

and Redirecting Number (RDN), as defined in ITU-T Q.732.2.

Using a testing tool:

a) R, H and C all belong to the HPMN
b) R is roaming on the VPMN network while H and C are located in the HPMN
c) A late call forward takes places and is not optimally routed (that is, the calls is
effectively routed towards the R destination and returned to the HPMN).

Computation

Abstract formula:

Number of Call Forwards including OCN/RDN

OCN & RDN [%] =  100%
Number of Call Forward tests
Trigger points:
OCN & RDN are correctly transmitted.

2.4.13 Call Completion Ratio Circuit Switched Telephony (CCR-CS-T) [2]

Abstract definition
Probability that a successful call attempt is maintained for a predetermined time until it is
released intentionally by R- or H-party.

See ETSI TS 102 250-2 (6.6.5 Cut-off Call Ratio).

See also complementary QoS Indicator: Call Non-Completion Rate circuit switched
(CNCR-CS).

Computation

Abstract formula:
Number of intentionally terminated telephony calls
CCR - CS - T [%] = 100%
Number of successful telephony call attempts

Trigger points:
Successful call attempt: Connect measurement (alerting’ tone or
ALERTING message detected by R-party)

Terminated call: Release of connection directly by R- or H-party

V0.1 Page 31 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.4.14 ALOC (Average Length of a Call) [4]

Abstract Definition
As defined in ITU E-437: average duration of calls. The advice is to measure this for MOC
and MTC separately, as there could be a significant natural difference between these 2 call
types.

Computation
Using traffic report:

Total network usage of MOC seconds in a month

ALOC MOC [seconds] =
Total number of MOC calls in a month

Total network usage of MTC seconds in a month

ALOC MTC [seconds] =
Total number of MTC calls in a month

2.4.15 CLI Transparency [5]

Call Line Identification (CLI) between countries is often not transmitted, the display indicating
“PRIVATE”, "UNAVAILABLE" or "INTERNATIONAL". This is usually the case with different
network types and international roaming scenarios.

Overseas number may be compressed into a "domestic" format and thus possibly not be
recognizable: e.g. a US number <1 555 555 7878> may be displayed in the UK as <555 555
7878>, instead of <001 555 555 7878> (or as +1 555 555 7878), where the "+" represents
the access code to dial international numbers).

Abstract Definition
CLI needs to be delivered correctly and complete, in a way it can be used to dial back the
original called party.

Computation
Number of complete and correct CLI' s
CLI transparency [%] = *100%
Number of calls

2.4.16 Speech Quality on Call Basis (SpQ) [2] [21] [27] [28] [29]Abstract
definition
Telephony speech quality on call basis is an indicator representing the quantification of the
end-to-end speech transmission quality of the Mobile Telephony Service. This parameter
computes the speech quality on the basis of completed calls7.

Computation
The validation of the end-to-end quality is made using MOS-LQO scales. These scales
describe the opinion of users with speech transmission and its troubles (noise, robot voice,
echo, dropouts and so on), according to ITU-T Recommendation P.862 PESQ in conjunction

7 The acoustic behaviour of mobile terminals is not part of this speech quality measurement.

V0.1 Page 32 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

with ITU-T Recommendation P.862.1, or according to ITU-T Recommendation P.863

POLQA. The algorithm used should be reported. The speech quality measurement is taken
per call. An aggregation should be made on one value for speech quality per call.

Abstract formula:
CS, CSFB

Telephony Speech Quality on Call Basis (received R - party) = f (MOS - LQO )

Telephony Speech Quality on Call Basis (received H - party) = f (MOS - LQO )

Optionally it might be useful to aggregate both speech quality values into one. In this case
the worst of both shall be used. This aggregated speech quality value shall be called SpQ
(min).

Trigger points:
Beginning of connection: Interchange speech samples between R-party
and H-party

End of connection: Release of connection

Note: The acoustic behaviour of terminals is not part of this speech quality
measurement.

2.4.18 CSFB Return to LTE Success Ratio

Abstract definition
This parameter measures the probability that a UE has to re-join the LTE network after a
CSFB call within a pre-determined time interval.

For a valid calculation the following preconditions must be met:

• LTE coverage is present at the end of the CSFB call

• A call is successfully established (with or without CSFB at H-Party)
• The call is regularly disconnected (i.e. no drop)

Computation

Abstract formula

successful Return to LTE attempts

Return to LTE Success Ratio [%] =  100
all Return to LTE attempts

2.4.19 Trigger points

Start: Hang up the call

V0.1 Page 33 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Stop: network type indicator on the UE’s display switches to LTE8Remarks

This KPI shall be separately tested for CSFB voice MO call and MT call.

2.4.20 CSFB Return to LTE TimeAbstract definition

This parameter measures the time needed by the UE to re-join the LTE network after a
CSFB call.

Computation

Abstract formula

Return to LTE Time [s] = (tfirst SIB message in LTE received − t call disconnect ed )[s]

Trigger points
tcall disconnect ed time when the call is disconnected
tfirst SIB message in LTE received time when the first SIB message in LTE is received

Remarks
This KPI shall be separately tested for CSFB voice MO call and MT call.

2.5 Short Message Service

The SMS KPIs are applied to SMS over GSM, UMTS CS or SMSoSGs.

2.5.1 Service Accessibility SMS MO (SA SMS MO) [2]

Abstract definition
Probability that the end-user can access the Short Message Service (SMS) when requested
while it is offered by display of the network indicator on the Mobile Equipment. In this case
the user wants to send a Short Message.

See ETSI TS 102 250-2 (7.4.2 SMS Service non-accessibility).

Computation
Note: For the trigger point explained here, the connection over the air interface must be
measured (for example Layer-3) and the answers of the SMSC must be counted statistically.
The protocol for every connection shows the deviation from the successful service access.

Only the first try should be measured. If the Short Message is established with the second
try this should not be counted.

8In a measurement system with automatic call dialling this KPI is influenced by the call interval. In
particular the call interval will implicitly set the timeout for Return to LTE calculation.

V0.1 Page 34 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula:
Number of successful SMS service attempts
Service Accessibility SMS MO [%] =
Number of all SMS service attempts

Trigger points [for example Layer-3 messages]:

Start SMS service attempt: Initiate sending an SMS at Roaming side

Successful SMS service Receiving acknowledgement of the SMSC

attempt:

2.5.2 Service Accessibility SMS MT (SA SMS MT) [2]Abstract definition

Probability that the end-user can receive a Short Message from its Home Network SMS-C
while it is offered by display of the network indicator on the Mobile Equipment. In this case
the user wants to receive a Short Message.

See ETSI TS 102 250-2 (7.4.2 SMS Service non-accessibility).

Computation
Only the valid attempts have to be measured. Errors due to user mistake (for example
memory full) should be excluded.

Abstract formula:
Number of successful SMS - MT service attempts
Service Accessibility SMS MT [%] =
Number of all SMS - MT service attempts

Trigger points [for example Layer-3 messages]:

Start SMS service attempt: Initiate sending a SMS from Home SMS-C

Successful SMS service Receiving Short Message at Roaming side

attempt:

2.5.4 Access Delay SMS MO (AD SMS-MO) [2]Abstract definition

Time between sending a Short Message to a Short Message Centre and receiving the
notification from the Short Message Centre.

See ETSI TS 102 250-2 (7.4.3 SMS Access Delay).

Computation

Abstract formula:
Access Delay SMS MO [s] = t receive − t send SMS

treceive: point of time the mobile equipment receives the send confirmation from the SMS
Centre

tsend SMS: point of time the user sends his SMS to the SMS Centre

V0.1 Page 35 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Trigger points [for example Layer-3 messages]:

Start SMS service attempt: Initiate sending an SMS at Roaming side

Successful SMS service Receiving acknowledgement of the SMSC

attempt:

2.5.6 Access Delay SMS MT (AD SMS-MT) [2]Abstract definition

Time between sending a Short Message from the Home Short Message Centre and
receiving the notification at the Short Message Centre.

See ETSI TS 102 250-2 (7.4.3 SMS Access Delay).

Computation

Abstract formula:

Access Delay SMS MT [s] = (t receive - t send SMS )[s]

treceive: point of time the SMS Centre receives confirmation that the Short Message was
correctly delivered.

tsend SMS: point of time the Short Message leaves the SMS Centre

Trigger points [for example Layer-3 messages]:

Start SMS service attempt: Initiate sending a SMS from Home SMS-C

Successful SMS service Receiving Short Message Confirmation of

attempt: Delivery at roaming side

2.5.8 End-to-End Delivery Time for SMS MO (E2E DT SMS-MO) [2]Abstract

definition
The SMS end-to-end delivery time is the time between sending a short message from R-
party in VPMN to a Short Message Centre and H-party in HPMN receiving the very same
short message from the Short Message Centre.

See ETSI TS 102 250-2 (7.4.5 End-to-End Delivery Time).

Computation

Abstract formula:

SMS MO End - to - End Delivery Time [s] = (t H, receive - t R, send )[s]

V0.1 Page 36 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

t H, receive : point of time the H-party in the HPMN receives the short message from the SMS
Centre.

t R, send : point of time the user sends the SMS to the SMS Centre.

Remarks:
Not relevant for QoS Roaming SLA since time measured is dependent on the performance
of the HPMN SMS-C.

End-to-End Delivery Time for SMS MT (E2E DT SMS-MT) [2]Abstract definition

The SMS end-to-end delivery time is the time between H-party in HPMN sending a short
message to a Short Message Centre and R-party in VPMN receiving the very same short
message from the Short Message Centre.
See ETSI TS 102 250-2 (7.4.5 End-to-End Delivery Time).

Computation

Abstract formula:

SMS MT End - to - End Delivery Time [s] = (t R, receive - t H, send )[s]

t R, receive : point of time the R-party in the VPMN receives the new short message from the
SMS Centre.

t H, send : point of time the H-party sends a short message to the SMS Centre in the HPMN.

Remarks:
Not relevant for QoS Roaming SLA since time measured is dependent on the performance
of the HPMN SMS-C.

2.5.10 Completion Ratio SMS (CR SMS) [2] [15]Abstract definition

Ratio of received and send Test SMS from one mobile to another mobile part, excluding
duplicate received and corrupted Test SMS.

A corrupted Test SMS is a SMS with at least one bit error.

For test and measurement purposes a message is considered valid if it is delivered

successfully within a time window defined.

See ETSI TS 102 250-2 (7.4.4 SMS Completion Failure Ratio)

V0.1 Page 37 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation

Abstract formula:
successful received Test SMS - duplicate received Test SMS - corrupted Test SMS
CR SMS CS [%] =
Number of all send Test SMS

Trigger points:
Successfully send and received SMS via SMSC.

Time window of measurements according to user profile.

2.6 Circuit Switched Data Service

2.6.1 Service Accessibility, Circuit Switched Data (SA –CSD) [15]

Abstract definition
Probability that the end-user’s DTE can access the Mobile Data Service when requested.
This will be indicated by the DTE receiving the valid ‘connect’ message from the distant
DTE.

Probability that the end-user’s DTE can access the Mobile Data Service when requested.

There are 2 layers of accessibility for CSD

6. Access to the target network DCE.

7. Access to the required data service provided by a data server.
To a user, these 2 events would be seamless and therefore the calculation for the service
access should be a composite of these 2 activities. The field test system therefore must
automate and combine the two layers to provide a single SA-CSD metric.

To combine the 2 layers should involve calculation of the success of the following actions.

8. ATDT command including target number.

9. Receive Connect from target network DCE
10. Send relevant command to target Data Server.
11. Receive valid response from Data Server
The specific commands and responses from data servers will be detailed in ‘Typical
procedures for quality of service measurement equipment’.

Computation
A successful call attempt is when the A-party DTE receives valid response from test server.
This can either be a dedicated data test server or a data server accessed when testing
functionality via the public internet.

V0.1 Page 38 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula:
Number of successful call attempts
Service Accessibility CSD =
Number of call attempts

Trigger points:
Beginning of the call ATDT command with dialled number sent by A-
attempt: party DTE.

Successful call attempt: Valid response received from Data Server.

2.6.2 Set-up Time (ST – CSD)Abstract definition

Time between sending of complete address information in ATDT command by A-Party and
receipt of valid response from data server.

Computation

Abstract formula:
Set - up Time Circuit Switched Data [s] = t 2 - t 1

t1: point of time where A-party DTE sends ATDT command

t2: point of time where connect is established (valid response received by A-party from data
server)

Trigger points:
Beginning of the Set-up Sending of ATDT command by A-party
time measurement:

Successful connection: Valid response received from Data Server.

2.6.4 Data Quality (DQ-CSD)

For definitions of Data Quality Parameters refer to section 2.7.

Completion Ratio Circuit Switched Data (CR-CSD)

Abstract definition
Probability that a successful call attempt is not released except when intended by any of the
parties involved in the call.

V0.1 Page 39 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation

Abstract formula:
Number of calls terminated by end users
Call completion Ratio CSD =
Number of successful data call attempts

Trigger points:
Successful call attempt: Valid response received by A-party DTE.

Completed call: DTE ‘ready’ only when call ended by either party
intentionally.

2.7 Packet Switched Data Service (General Packet Radio Service)

For test purposes it will be necessary to have the mobile test equipment in a stable state
before testing. For each test the mobile should begin by being powered on and attached but
not PDP context activated. Specific details are to be found in ‘Typical procedures for quality
of service measurements’.

Note: The bearer technology will affect the monitoring results for many of the Packet
Switched Data measurements.

2.7.1 Service Accessibility Ratio – Packet Switched Data (SA – PSD) [2] [7]

Abstract definition
Probability that a subscriber can successfully attach to the PS network. As defined in ETSI
TS 129 002. See also ETSI TS 102 250-2 (5.3 Attach Failure ratio).

There are 2 layers of accessibility for GPRS:

• Access to the mobile network GPRS core infrastructure.

• Access to the required data service provided by a data server.
To a user, these 2 events would be seamless and therefore the calculation for the service
access should be a composite of these 2 activities. The field test system therefore must
automate and combine the two layers to provide a single SA-PSD metric.

To combine the 2 layers should involve calculation of the success of the following actions.

•
Sending of valid command (for example ATD*99# (with IP address of target server))
from A party DTE to obtain IP connection.
• Receive valid response from GGSN
• Send valid command to target Data server.
• Receive valid response from target Data server.
The specific commands and responses from data servers will be detailed in ‘Typical
procedures for quality of service measurements’.

If multiple Access Point Names (APNs) are used, the measures should be performed on a
per-APN basis for consistency in measurements.

V0.1 Page 40 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation
A session will be considered set-up successfully if a valid response is received from the
target data server

Abstract formula:
Number of successful session attempts
Service Accessibility PSD =
Number of session attempts

Trigger points:
Beginning of the session Send valid command request (for example
attempt: ATD*99#
(with IP address of target server))

Successful session attempt: Valid response received from target data server

2.7.2 Set-up Time – Packet Switched Data (ST – PSD) [2]Abstract definition
Time between sending of valid command (for example ATD*99# (with IP address of target
server) message and receipt of valid response message from target data server.

Computation
A session will be considered set-up successfully if a valid response is received from the
target data server

Abstract formula:
Set - up Time Packet Switched Data [s] = t 2 - t 1

t1: point of time where A-party valid session request command

t2: point of time where connect is established (valid response received by A-party from data
server)

Trigger points:
Beginning of the session Send valid command request (for example
attempt: ATD*99#
(with IP address of target server))

Successful session attempt: Valid response received from target data server

Note for all data quality testing it is assumed that for each test, PDP Context is activated
and at the end of the individual test PDP Context is de-activated.

For definitions of Data Quality Parameters refer to section 2.7.

V0.1 Page 41 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.7.4 Service Integrity - Throughput (Kbit/sec) [2]

Abstract definition
This parameter describes the average data transfer rate at the network transport level (and
not at the User Application level), based on the Mean Data Rate as defined by ETSI TS 102
250-2 (6.1.7 Mean Data Rate).

The prerequisite for this parameter is network and service access.

Computation

Abstract formula:
VolumeOfDataTransferred
BitPipeThroughput =
TransferTime

Remarks:
The measurement of Throughput will be influenced by Packet Loss and Roundtrip Time
(Delay). Throughput measurements may also be influenced by service-side factors such as
radio cell reservation and network usage. Mobile Station ()

2.7.5 Service Integrity - Goodput (Kbit/sec) [10]Abstract definition

This parameter describes the average data transfer rate at the User Application level (and not
at the network transport level).
The prerequisite for this parameter is network and service access.
Goodput is defined in IETF RFC2647.

Computation

Abstract formula:
Goodput may be calculated as:

VolumeOfUsefulDataTransferred
BitPipeGoodput =
TransferTime

Please note that the definition of “useful data” depends on the user applications used for the
measurement.

Remarks:
The measurement of Goodput will be influenced by Packet Loss and Roundtrip Time
(Delay). Goodput measurements may also be influenced by service-side factors such as
packet size and the User Application.

2.7.7 Service Integrity - Roundtrip Time [11]Abstract definition

Roundtrip Time (Roundtrip Delay) is the total time that it takes to transmit an IP packet from
the source to the destination and receive the reply packet from the destination at the source.

The prerequisite for this parameter is network and service access.

V0.1 Page 42 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

See ‘Delay’ in section 8 of PRD IR.34 ‘Inter-Service Provider IP Backbone Guidelines’.

Computation

Abstract formula:

Round Trip Time (ms)=(timestamp Packet received)- (timestamp Packet sent)

2.7.9 Service Integrity – Packet Loss [6] [11]

Abstract definition
Packet Loss is the ratio of dropped packets to all packets sent from the source to Destination
over a given period of time.

The prerequisite for this parameter is network and service access.

See ‘Packet Loss Ratio’ in section 8 of PRD IR.34 ‘Inter-Service Provider IP Backbone
Guidelines’.

2.7.10 Completed Session Ratio (CoSeR – PSD)Abstract definition

Probability that a successful session attempt is not released for a reason other than
intentional by any of the parties involved in the session.

Computation

Abstract formula:
Number of sessions not released other than by end user
Completed Session Ratio PSD = 
Number of successful data session attempts

Trigger points:
Successful session attempt: Valid response received from target data server.

V0.1 Page 43 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Completed session: Session released intentionally by either end-

user.

2.7.12 Service Retainability – Average PDP Context Average Session Time

[8]Abstract definition
The average PDP context session time is the average duration of the PDP context sessions
successfully completed.

PDP Context is defined in 3GPP 29.060.

Computation

Abstract formula:

PDPContext AverageSessionTime = Average(t PDPContextDeactivation − t PDPContextActivation )

a. Trigger points:
Start: Notification of successful PDP context activation
(Activate PDP Context Accept)

End: PDP context deactivation request initiated by the

user (Deactivate PDP Context Request)

Remarks:
The PS bearer has to be active in the cell used by a subscriber (cf. Unavailability) and the
mobile station has to be attached as well as the respective PDP context has to be activated.

2.8 Data Service Class Definitions and Measurements

The following definitions for data services and data quality DQ are relevant for both circuit
switched and packet switched data as, the different classes of data service will be applied
identically irrespective of the data bearer system.

Note that data quality will be a result of an overall call or session. For test purposes it may
be desirable to break this down into geographically distinct measurements but for QoS
reporting should be kept to call or session lengths.

Data classes are defined in 3rd Generation Partnership Project Technical Specification
Group Services and System Aspects; QoS Concept and Architecture (3G TR 23.907) - see
Table 1.

V0.1 Page 44 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Traffic class Conversational Streaming class Interactive class Background

class streaming RT Interactive best Background
conversational RT effort best effort
Fundamental Preserve time Preserve time Request response Destination is
relation (variation) relation (variation) pattern not expecting
characteristics between information between Preserve payload the data within a
entities of the stream information entities content certain time
Conversational of the stream Preserve
pattern (stringent and payload content
low delay )
Example of the Voice Streaming video Web browsing Background
download of
application emails
Table 2: UMTS QoS classes

2.8.1 Conversational Class Data

Abstract definition
Indicator representing the end-to-end data transmission quality of the Conversational Class
Data Service. This represents full duplex transfer of data in near real time.

Computation
The end-to-end data quality is validated by measuring the average data throughput in both
up-link and down link direction on a best effort basis. The data throughput measurement will
be computed and averaged over the duration of the session/call and reported in bits per
second. Additionally the minimum throughput averaged over 10% of the overall call/session
length, the maximum throughput over 10% of the overall call/session length and worst. The
worst delay time for the call/session should also be reported

2.8.2 Abstract formula:

DQ(receive d A - side) = X bits/sec

DQ(receive d B - side) = X bits/sec

2.8.3 Trigger points:

Beginning of call/session Interchange data frames of predefined data
data sample: between A and B-party DTE

End of call/session data Calculation of average data throughput for

sample: Call/session data sample

2.8.4 Streaming ClassAbstract definition

Indicator representing the end-to-end data transmission quality of the Mobile, Circuit
Switched, Streaming Class Data Service. This measure represents a delivery of data in one
direction (up-link or down-link) in near real time for example video broadcast.

V0.1 Page 45 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Additionally the minimum throughput averaged over 10% of the call/session duration, the
maximum throughput averaged over 10% of the call/session duration and the worst block
error rate. The worst delay time for the call/session should also be reported

Note for streaming class service only the down link direction is considered, but if service
applications are introduced for uplink streaming then this can be added for calculation for
data received by B-Party

Computation
The end-to-end data quality is validated by measuring the data throughput in down link
direction on a best effort basis. The data throughput measurement will be computed and
averaged over the duration of the call/session and be reported in bits/sec.

Abstract formula:
DQ(receive d A - side) = X bits/sec

Trigger points:
Beginning of Call/session Transmission of data frames of indexed
data sample: predefined data B-party to A-party

End of Call/session data Calculation of average data throughput for

sample: call/session data sample

2.8.6 Interactive ClassAbstract definition

Indicator representing the end-to-end data transmission quality of the Mobile Circuit
Switched Interactive Class Data Service. This represents duplex transfer of data in non real-
time.

Computation
The validation of the end-to-end data quality is made by the time taken to download
specified files of fixed data size to the A-party DTE when, requested by the A-party sending
a request to the data server.

Assumption: The A-party DTE has already been connected to the data server as part of the
call set-up process.

Abstract formula:
DQ download time [s] = t 2 - t 1

t1: point of time where A-party DTE sends data request.

t2: point of time where A-party receives complete uncorrupted requested file/s

V0.1 Page 46 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Trigger points:
Beginning of request for Data request sent by A-party DTE
download:

Download of file/s complete: Uncorrupted file/s received by A-party DTE.

2.8.8 Background classAbstract definition

Indicator representing the end-to-end data transmission quality of the Mobile Circuit
Switched Background Class Data Service. This represents data transfer with no real-time
dependency (although for QoS testing, data transfer time is measured).

Computation
The validation of the end-to-end data quality is made by the time taken to download a file/s
of fixed data size to the A-party DTE when, requested by the A-party sending a request to
the target server.

Assumption: The A-party DTE has already been connected to the data server as part of call
set-up process.

Abstract formula:
DQ File download time [s] = t 2 - t 1

t1: point of time where A-party DTE sends data transfer request

t2: point of time where A-party receives complete uncorrupted file/s

Trigger points:
Beginning of request for Request sent by A-party DTE
download:

Download of file/s complete: Uncorrupted file/s received by A-party DTE

2.9 FTP QoS Parameters

Precondition: PS attached and the respective PDP context / default EPS bearer context
activated.

Assumption: a single TCP/IP connection is applied.

2.9.1 FTP {Download|Upload} IP Service Access Success Ratio [%] [2]

Abstract definition
The IP-service access ratio denotes the probability that a subscriber can establish a TCP/IP
connection to the server of a FTP service successfully.

See ETSI TS 102 250-2 (6.1.3 FTP IP-Service Failure Ratio)

V0.1 Page 47 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation

Abstract Formula

FTP {Download | Upload} IP - Service Access Success Ratio [%] =

successful attempts to establish an IP connection to the server
 100
all attempts to establish an IP connection to the server

Trigger Points
Download:
Start: Initiate file download, or the first [SYN] sent on the data socket.
Stop: File download starts, or
Method A9: Reception of the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.
Upload:
Start: Initiate file upload, or the first [SYN] sent on the data socket.
Stop: File upload starts, or
Method A: Sending the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.

2.9.2 FTP {Download|Upload} IP Service Access Setup Time [s] [2]Abstract

definition
The IP-service setup time denotes the time period needed to establish a TCP/IP connection
to the server of a FTP service, from sending the initial query to the server to the point of time
when the content is sent or received.

See ETSI TS 102 250-2 (6.1.4 FTP IP-Service Setup Time)

Computation

Abstract Formula

FTP {Download | Upload} IP - Service Setup Time [s] = (t IP -Service access successful - t IP -Service access start )[s]

Trigger Points
Download:
Start: Initiate file download, or the first [SYN] sent on the data socket.
Stop: File download starts, or
Method A: Reception of the first data packet containing content

9Method A is used for payload throughput, method B is used for transaction throughput, ref. to ETSI
TS 102 250-2, 4.2 FTP, HTTP and E-mail issues.

V0.1 Page 48 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.
Upload:
Start: Initiate file upload, or the first [SYN] sent on the data socket.
Stop: File upload starts, or
Method A: Sending the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.

2.9.4 FTP {Download|Upload} Session Success Ratio [%] [2]Abstract

definition
The session success ratio is the proportion of completed sessions and the total sessions that
were started successfully.

See ETSI TS 102 250-2 (6.1.5 FTP Session Failure Ratio)

Computation

Abstract Formula

completed sessions
FTP {Download | Upload} Session Success Ratio [%] =  100
successfully started sessions

Trigger Points
Download:
Start: Initiate file download, or the first [SYN] sent on the control socket.
Stop: File download successfully completed, or Reception of the last data packet
containing content
Upload:
Start: Initiate file upload, or the first [SYN] sent on the control socket.
Stop: File upload successfully completed, or Reception of the [FIN, ACK] for the
last data packet containing content.

2.9.6 FTP {Download|Upload} Session Time [s] [2]Abstract definition

The session time is the time period needed to successfully complete a file transfer session.

See ETSI TS 102 250-2 (6.1.6 FTP Session Failure Ratio)

Computation

Abstract Formula

FTP {Download | Upload} Session Time [s] = (t session end - t session start )[s]

Trigger Points
Download:

V0.1 Page 49 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Start: Initiate file download, or the first [SYN] sent on the control socket.
Stop: File download successfully completed, or Reception of the last data packet
containing content
Upload:
Start: Initiate file upload, or the first [SYN] sent on the control socket.
Stop: File upload successfully completed, or Reception of the [FIN, ACK] for the
last data packet containing content.

2.9.8 FTP {Download|Upload} Mean Data Rate [kbit/s] [2]Abstract Formula

After a data link has been successfully established, this parameter describes the average
data transfer rate measured over the entire connect time to the service. The data transfer
shall be successfully terminated. The prerequisite for this parameter is network and service
access.

See ETSI TS 102 250-2 (6.1.7 FTP Mean Data Rate)

Computation

Abstract Formula

user data transferr ed [kbit]

FTP {Download | Upload} Mean Data Rate [kbit/s] =
( )
t data transfer complete - t data transfer start [s]

Trigger Points
The average throughput is measured from opening the data connection to the end
of the successful transfer of the content (file).
Download:
Start: File download starts, or
Method A: Reception of the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket
Stop: File download successfully completed, or Reception of the last data packet
containing content.
Upload:
Start: Fie upload starts, or
Method A: Sending of the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.
Stop: File upload successfully completed, or Reception of the [FIN, ACK] for the
last data packet containing content.

2.9.10 FTP {Download|Upload} Data Transfer Success Ratio [%] [2]Abstract

definition
The data transfer success ratio is the proportion of completed data transfers and data
transfers that were started successfully.

V0.1 Page 50 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

See ETSI TS 102 250-2 (6.1.8 FTP Data Transfer Cut-off Ratio)

Computation

Abstract Formula

FTP {Download | Upload} Data Transfer Success Ratio [%] =

completed data transfers
 100
successfully started data transfers

Trigger Points
Download:
Start: File download starts, or
Method A: Reception of the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.
Stop: File download successfully completed, or Reception of the last data packet
containing content
Upload:
Start: File upload starts, or
Method A: Sending of the first data packet containing content
Method B: Reception of the [ACK] from the [SYN, ACK] for active mode
connections, sending of the [ACK] for the [SYN, ACK] for passive mode
connections on the data socket.

Stop: File upload successfully completed, or Reception of the [FIN, ACK] for the
last data packet containing content.

2.9.12 FTP {Download|Upload} Data CapacityAbstract definition

FTP {Download|Upload} denotes the maximum FTP download or upload Bandwidth of data
capacity.

Computation

Abstract Formula

n
FTP {Download | Upload} Bandwidth [mbit/s] =  DataRateOfEachTCPconnection
1

Trigger points
Start: multiple TCP/IP connections are established and download (or upload) one or multiple
FTP files from (or to) one or multiple different FTP servers in parallel.

Stop: if all FTP files are downloaded (or uploaded) or a fixed time duration is reached.

V0.1 Page 51 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Remarks
The test purpose is to measure the available download or upload Bandwidth of the network
in case of PS data roaming. The FTP data contents and the establish time of the multiple
TCP/IP connections are irrelevant.

2.10 WEB Browsing (HTTP / HTTPS) QoS Parameters

Preconditions: PS attached and the respective PDP context / default EPS bearer context
activated.

The KPIs defined in this section are generic and are applicable to HTTP or HTTPS.

Assumption: a single TCP/IP connection.

2.10.1 HTTP IP-Service Access Success Ratio [%] [2]

Abstract definition
The IP-service access ratio denotes the probability that a subscriber can establish a TCP/IP
connection to the server of the HTTP service successfully.

See ETSI TS 102 250-2 (6.8.3 HTTP IP-Service Access Failure Ratio)

Computation

Abstract Formula

HTTP IP - Service Access Success Ratio [%] =

successful attempts to establish an IP connection to the server
 100
all attempts to establish an IP connection to the server

Trigger points
Start: enter the URL and hit “Return” or First [SYN] sent.
Stop: Web page download starts, or
Method A: Reception of the first data packet containing content.
Method B: Sending of the first GET command.

2.10.2 HTTP IP-Service Setup Time [s] [2]Abstract definition

The IP-service setup time is the time period needed to establish a TCP/IP connection to the
server of a HTTP service, from sending the initial query to the server to the point of time
when the content is received.

See ETSI TS 102 250-2 (6.8.4 HTTP IP-Service Access Time)

Computation

Abstract Formula

HTTP IP - Service Setup Time [s] = (t IP -Service access successful - t IP -Service access start ) [s]

V0.1 Page 52 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

2.10.4 HTTP Session Success Ratio [%] [2]Abstract definition

The completed session ratio is the proportion of completed sessions and sessions that were
started successfully.

See ETSI TS 102 250-2 (6.8.5 HTTP Session Success Failure Ratio)

Computation

Abstract Formula

completed sessions
HTTP Session Success Ratio [%] = 100
successfully started sessions

Trigger points
Start: enter the URL and hit “Return” or first [SYN] sent.
Stop: The complete Web page appears in the browser window, or Reception of
the last data packet containing content.

2.10.6 HTTP Session Time [s] [2]Abstract definition

The session time is the time period needed to successfully complete a HTTP session.

See ETSI TS 102 250-2 (6.8.6 HTTP Session Time)

Computation

Abstract Formula

HTTP Session Time [s] = (t session end - t session start )[s]

Trigger points
Start: enter the URL and hit “Return” or first [SYN] sent.
Stop: The complete Web page appears in the browser window, or Reception of
the last data packet containing content

2.10.8 HTTP Mean Data Rate [kbit/s] [2]Abstract definition

See ETSI TS 102 250-2 (6.8.7 HTTP Mean Data Rate)

V0.1 Page 53 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Computation

Abstract Formula

user data transferr ed [kbit]

HTTP Mean Data Rate [kbit/s] =
( )
t data transfer complete - t data transfer start [s]

Trigger points
Start: Web page download starts or First [SYN] sent, or
Method A: Reception of the first data packet containing content.
Method B: Sending of the first GET command
Stop: Web page download successfully completed, or reception of the last data
packet containing content

2.10.10 HTTP Data Transfer Success Ratio [%] [2]Abstract definition

The data transfer success ratio is the proportion of completed data transfers and data
transfers that were started successfully.

See ETSI TS 102 250-2 (6.8.8 HTTP Data Transfer Cut-off Ratio)

Computation

Abstract Formula

completed data transfers

HTTP Data Transfer Success Ratio [%] = 100
successfully started data transfers

Trigger points
Start: Web page download starts or First [SYN] sent, or
Stop: Web page download successfully completed, or reception of the last data
packet containing content.

2.10.12 HTTP Content Compression Ratio [%] [2]Abstract definition

The HTTP content compression ratio denotes the compression level of the received data
accessible by the user agent in relation to the data sent by the origin server by using HTTP.
It takes into account the overall effects of loss and lossless compression and non-reversible
modifications of the original stored content during transmission.

See ETSI TS 102 250-2 (6.8.9 HTTP Content Compression Ratio)

Computation

Abstract Formula

 received HTTP content size 

HTTP Content Compression Ratio [%] = 1 -   100
 sent HTTP content size 

V0.1 Page 54 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Remarks
Regarding the download of images the HTTP content compression ratio gives no indication
on the quality of the compressed images as perceived by the user. The explanations on the
influence of performance enhancement proxies should be taken into account. The current
definition is applied to transferring HTTP content that consists of multiple objects (e.g. a web
page) or content that consists of a single object. The “sent HTTP content” is an external
input parameter for the calculation. It may be a constant (e.g. reference web page) or it may
be measured directly at the HTTP server during the test execution, in a different network.

2.10.14 HTTP {Download|Upload} Data CapacityAbstract definition

HTTP {Download|Upload} denotes the maximum HTTP download or upload Bandwidth of
data capacity.

Computation

Abstract Formula

n
HTTP {Download | Upload} Bandwidth [mbit/s] =  DataRateOfEachTCPconnection
1

Trigger points
Start: multiple TCP/IP connections are established and download (or upload) one or multiple
HTTP files from (or to) one or multiple different HTTP servers in parallel.

Stop: if all HTTP files are downloaded (or uploaded) or a fixed duration is reached.

Remarks
The test purpose is to measure the available download or upload Bandwidth of the network
in case of PS data roaming. The HTTP data contents and the establish time of the multiple
TCP/IP connections are irrelevant.

2.11 PING QoS Parameters

2.11.1 PING Packet Loss Ratio

Abstract definition
PING packet Loss ratio is the proportion of dropped PING packets vs. the total PING
packets sent.

Succssful PING packets received

PING Packet Loss Ratio [%] = (1 − )  100
Total number of PING packets sent

Trigger Points
Start: ICMP echo request sent,

Stop: ICMP echo reply received.

Remarks
An optional parameter is the number of PING packets sent.

2.11.2 PING Round Trip Time [ms] [2]Abstract definition

The round trip time is the time required for a packet to travel from a source to a destination
and back. It is used to measure the delay on a network at a given time. For this
measurement the IP connection must already be established.

See ETSI TS 102 250-2 (6.3.1 PING roundtrip time)

Computation

Abstract Formula

( )
Ping Round Trip Time [ms] = t packet received - t packet sent [ms]

Trigger Points
Start: ICMP echo request sent,

Stop: ICMP echo reply received by the sender.

Remarks
The size of the PING data packet is an optional parameter which will impact on the result of
the measurement.

3 QoS Parameters for Interconnection

The QoS parameters (KPIs) specified in this section are mainly used for TDM
Interconnection between a Mobile Operator and an International Carrier for International
Transit Voice.

3.1 ASR = (Answer Seizure Ratio)

ITU E.425 [14] ASR represents the relationship between the number of seizures that result
in an answer signal and the total number of seizures.

ASR is measured by the mobile operator on outgoing traffic on their gateway switch.

V0.1 Page 56 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

3.2 ABR = (Answer BID Ratio)

ITU E.425 [14] ABR represents the relationship between the number of bids that result in an
answer signal and the total number of bids.

ABR is to be measured by the carrier on incoming traffic on their gateway switch.

3.3 CLI = (Calling Line Identification)

Calling Line Identification, also known as “A number” is basic information contained in the
signalling system that identifies the calling party.

CLIP – CLI Presentation provides for the calling user number to be displayed to the called
user.

CLIR – CLI Restriction provides a means for the calling user to restrict presentation of its
MSISDN to the called user.

A more detailed definition is referred to in section 2.3.11.

3.4 NER = (Network Efficiency Ratio)

ITU E.425 [14] NER expresses the relationship between the number of seizures and the sum
of the seizures resulting in an answer signal OR busy signal OR no answer. It excludes the
effects of customer behaviour and terminal behaviour. It represents the ability of the network
to deliver calls to the far end terminal.

A more detailed definition is referred to in sections 2.3.1 and 2.3.2.

3.5 PGAD = (Post Gateway Delay)

ITU E.437 [4] PGAD is the time interval between the seizure of the international circuit and
the receipt of the answer supervision.

3.6 PDD = (Post Dial Delay)

ITU E.431 [13] PDD is the time interval between dialling completion and the call connection
(ringing tone).

A more detailed definition is provided in sections 2.3.3 and 2.3.4.

3.7 ALOC = (Average Length of Conversation)

ITU E.437 [4] ALOC measures the average duration of calls. A statistically significant
difference in ALOC to the same destination on different routes may be investigated.

A more detailed definition is referred to in section 2.3.10.

3.8 Speech Quality on Sample Basis.

4 Reference: ITU-T Recommendation P.863 [27] VoLTE / ViLTE

Quality Parameters
This section contains VoLTE [31] and ViLTE [32] quality KPIs for interconnect and roaming
[18] [30] [33]. The quality parameters (KPIs) are divided into seven groups:

V0.1 Page 57 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

• Accessibility (IMS Registration quality)

• Voice / video service integrity and retainability
• SRVCC (PS-CS) quality
• EPS bearer quality
• Media transport quality
• Diameter quality10
The first three groups contribute to the service quality over LTE. The last three groups
characterize the quality of networks / transport.

4.1 VoLTE / ViLTE Service Quality Parameters

VoLTE / ViLTE service QoS parameters (KPIs) contains the service accessibility, integrity
mobility and preservation.

4.1.1 VoLTE / ViLTE Accessibility Parameters

IMS Registration Success Ratio

The SIP Registration procedure in IR.92 [31] and IR.94 [32] shall be followed. See also 5.3.2
of ETSI TR 103 219 [34].

Abstract definition
This parameter denotes the probability that UE successfully registers to IMS.

IMS Registration Success Ratio [%] =

 successful IMS Registration attempts  100
 all IMS Registration attempts

IMS Registration Time

The SIP Registration procedure in IR.92 [31] and IR.94 [32] shall be followed. See ETSI TR
103 219 [34] section 5.3.3.

Abstract formula

IMS Registration Time [s] = (tNetwork confirms IMS registrati on − tUE requests IMS Registrat ion )[s]

IMS Third-party Registration Success Ratio

Abstract definition
This parameter denotes the probability for the 3rd-party IMS registrations to be successfully
performed.

IMS 3rd - party registration success ratio [%] =

 successful IMS 3rd - party Registration attempts  100
 all IMS 3rd - partyRegistration attempts
10 The Diameter KPIs, denoting the quality of PCC and DRA quality, are not applied to e2e testing.

V0.1 Page 58 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

4.1.2 VoLTE / ViLTE Service Quality Parameters

The entire quality parameters in this section characterize the QoS aspect of the VoLTE /
ViLTE service integrity.

For the simplification of the technical description in a roaming scenario, H-party denotes a
party at HPMN, R-party denotes a Roaming party at the VPMN. In an interconnect scenario,
A-party denotes a party at HPMN A and B-party denotes a party at HPMN B where PMN A
and PMN B are interconnected.

For a valid KPI test, the following preconditions need to be met [34]:

• Default QCI5 EPS bearer context for the IMS well-known APN shall be established
and the EPS ATTACH message shall contain "IMS PS Voice" in the voice domain
preference list.
• IMS shall be present.
• LTE coverage shall be present.
• IMS registration shall be successfully accomplished.

VoLTE / ViLTE service access success ratio – NER MO (%)

Abstract definition
The KPI denotes the probability for the end-user to access the VoLTE / ViLTE service and
initiate a voice or video call. See also 2.3.1.1.

Abstract formula

VoLTE MO Accessibil ity [%] =

 Number of successful call attempts
 100
 Number of total call attempts

ViLTE MO Accessibility [%] =

 Number of successful call attempts
 100
 Number of total call attempts
Trigger points from an end user’s viewpoint, see 2.3.1.2..

Technical description

Start test: SIP "INVITE" sent by the R- / A-party.

Successfully stop test: ”180 Ringing” is received by the R- / A-party.

Unsuccessfully stop test: either the R- / A-party receives a 4xx error message as response
to the "INVITE", or timeout.

VoLTE / ViLTE service access success ratio – NER MT (%)

Abstract definition
The KPI denotes the probability that the end-user can access the VoLTE / ViLTE service and
receive a VoLTE / ViLTE call. See also 2.3.2.

V0.1 Page 59 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula

VoLTE MT Accessibil ity [%] =

 Number of successful call attempts
 100
 Number of total call attempts

ViLTE MT Accessibil ity [%] =

 Number of successful call attempts
 100
 Number of total call attempts
Trigger points from an end user’s viewpoint, see also 2.3.2.1.

Test start: Beginning of a call attempt of H-party / B-party, SIP "INVITE" received by the R-
Party / A-party.

Successfully Test stop : ”180 Ringing” is sent by the R-party / A-party and ”180 Ringing” is
received by H-party / B-party.

Unsuccessfully test stop : either the H-/ B-party receives a 4xx error message as response
to the "INVITE", or timeout.

VoLTE / ViLTE session setup time – PDD-MO (s)

Abstract definition
Time needed to setup an MO VoLTE / ViLTE call. See also 2.3.3.1.

Note: The KPI VoLTE / ViLTE session setup in this definition is referred to a
complete VoLTE / ViLTE session establishment, therefore is not applied to
SRVCC when voice session is handed over legacy technologies during the
voice session setup.

Abstract formula

VoLTE MO Session Setup Time [s] = (tCalling party receives notificati on − tCalling party initiates call session )[s]
ViLTE MO Session Setup Time [s] = (tCalling party receives notificati on − tCalling party initiates call session )[s]
Test start: R- / A- party initiates VoLTE session and sends SIP "INVITE"

Successful test stop : SIP: "200 OK (INVITE)" sent by H- / B-party and SIP: "200 OK
(INVITE)" received by R- / A-party [34]. In manual test, the time for the user accepting the
incoming call is excluded in the calculation.

Unsuccessful Test stop : R- / A-party receives a 4XX error message - the session set-up is
cancelled or does not receive any notification within a pre-determined time.

VoLTE / ViLTE session setup time – PDD-MT (s)

Abstract definition
Time needed to setup an MT VoLTE / ViLTE call. See also 2.3.4.1.

V0.1 Page 60 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Note: The KPI VoLTE / ViLTE session setup in this definition is referred to a
complete VoLTE session establishment, therefore is not applied to SRVCC
when voice session is handed over legacy technologies during the voice
session setup.

Abstract formula

VoLTE MT Session Setup Time [s] = (tCalling party receives notificati on − tCalling party initiates call session )[s]
ViLTE MT Session Setup Time [s] = (tCalling party receives notificati on − tCalling party initiates call session )[s]
Test start: H- / B-party initiates VoLTE / ViLTE session and sends SIP "INVITE" .

Successful test stop : SIP: "200 OK (INVITE)" sent by R- / A-party and "200 OK (INVITE)"
received by H- / B-party [34]. In manual test, the time for the user accepting the incoming call
is excluded in the calculation.

Unsuccessful test stop : H- / B-party receives a 4XX error message - the session set-up is
cancelled or does not receive any notification within a pre-determined time.

VoLTE / ViLTE session setup success ratio – CSSR-MO (%)

Abstract definition
VoLTE / ViLTE session setup success ratio denotes the probability that a successful MO
VoLTE / ViLTE call attempt results in a successfully established MO VoLTE / ViLTE call. See
also 3.2.5.

Abstract formula

VoLTE MO session setup ratio [%] =

 successful MO VoLTE calls  100
 all successful MO VoLTE call attempts
ViLTE MO session setup ratio [%] =
 successful MO VoLTE calls  100
 all successful MO VoLTE call attempts
Test start: R- / A-party initiates VoLTE / ViLTE session and sends SIP "INVITE"

Successful test stop : SIP "200 OK (INVITE)" sent by H- / B-party and SIP "200 OK
(INVITE)" received by R- / A-party [34]. In manual test, the time for the user accepting the
incoming call is excluded in the calculation.

Unsuccessful test stop : R- / A-party receives a 4XX error message - the session set-up is
cancelled or does not receive any notification within a pre-determined time. Any intentional
call terminations at R- / A-party or H- / B-party are excluded.

V0.1 Page 61 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

VoLTE / ViLTE session setup success ratio – CSSR-MT (%)

Abstract definition
VoLTE session setup success ratio denotes the probability that a successful MT VoLTE /
ViLTE call attempt results in a successfully established MT VoLTE / ViLTE call. See also
3.2.6.

Abstract formula

VoLTE MT session setup ratio [%] =

 successful MT VoLTE calls  100
 all successful MT VoLTE call attempts
ViLTE MT session setup ratio [%] =
 successful MT VoLTE calls  100
 all successful MT VoLTE call attempts
Test start: H- / B-party initiates VoLTE / ViLTE session and sends SIP "INVITE" .

Unsuccessful test stop: H- / B-party receives a 4XX error message that the session set-up is
cancelled or does not receive any notification within a pre-determined time. Any intentional
call terminations at R- / A-party or H- / B-party are excluded.

VoLTE / ViLTE session duration – ALOC-MO (s)

Abstract definition
MO VoLTE / ViLTE call duration, see also 2.3.10.

Abstract formula

VoLTE MO session duration [s] = (tBYE − t200 OK (INVITE) )[s]

ViLTE MO session duration [s] = (tBYE − t200 OK (INVITE) )[s]

An MO VoLTE / ViLTE session duration is the time measured at R- / A-party between
receiving "200 OK (INVITE)" and sending “BYE”. The call release is initiated by R- / A-party.

VoLTE / ViLTE session duration – ALOC-MT (s)

Abstract definition
MT VoLTE call duration, see also 2.3.10.

Abstract formula

VoLTE MT session duration [s] = (tBYE − t200 OK (INVITE) )[s]

V0.1 Page 62 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

ViLTE MT session duration [s] = (tBYE − t200 OK (INVITE) )[s]

An MT VoLTE / ViLTE session duration is the time measured at R- / A-party between
sending "200 OK (INVITE)" and receiving “BYE”. The call release is initiated by H- / B-party.

OIP transparency – CLI transparency

Abstract definition
VoLTE call originating identification presentation (OIP) needs to be delivered and presented
correctly and complete. It can be used to call back to the original called party. See also
2.3.11.

Abstract formula

OIP transparency ratio [%] =

 Number of complete and correct OIPs  100
 Number of successful VoLTE calls
VoLTE speech quality (SpQ MOS-LQO)
The VoLTE Speech Quality represents the end-to-end speech quality of the VoLTE service.

ETSI TS 102 250-2 [2] and defines two variants of a "Telephony Speech Quality" KPI. Both
are based on the same single measurement. A single measurement consist of the MOS-
LQO value determined for a single transferred speech sample, according to ITU-T P.863
POLQA [27]. A speech call usually consists of multiple speech samples transferred in both
directions.

The first variant, called "Telephony Speech Quality on Call Basis" (ETSI TS 102 250-2 [2]
clause 6.6.3) does a pre-aggregation of the measured MOS-LQO values to one value per
call (in either direction). ETSI TS 102 250-2 does not specify the aggregation method for this
pre-aggregation. A suitable method can be found in ETSI TR 102 506 [28].

The second variant, called "Telephony Speech Quality on Sample Basis" (ETSI TS 102 250-
2 clause 6.6.4) does not pre-aggregate the measured MOS-LQO values. In other words, one
call will result in multiple values for this KPI.

Averaging the results obtained in a measurement campaign with repeated calls will in
general yield different statistics for the two variants. Due to its pre-aggregation, the
"Telephony Speech Quality on Call Basis" focusses more on how users judge entire calls,
while the "Telephony Speech Quality on Sample Basis" gives insight into the overall speech
transmission quality for a particular connection or interconnection.

Abstract definition
VoLTE speech quality in roaming is measured on Call Basis, See also 2.3.12.

Abstract formula

VoLTE Speech Quality on Call Basis (received R - party) = f (MOS - LQO )

VoLTE Speech Quality on Call Basis (received H - party) = f (MOS - LQO )

V0.1 Page 63 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

The validation of the end-to-end quality is made using MOS-LQO scales [2]. These scales
describe the opinion of users with speech transmission and its troubles (noise, robot voice,
echo, dropouts, time scaling introduced by the jitter buffer, etc.) according to ITU-T P.863
[27]. The scale used has to be reported. An aggregation for measurement campaigns in
VoLTE roaming is made on call basis (the 1st variant).

The KPI shall be tested at the call terminating side and at the call originating side. The two
streams shall be successfully tested at the same time.

VoLTE speech quality (SpQ R-Factor-LQ)

Abstract definition
The R-Factor is an estimated speech quality rating defined by the E-Model (ITU-T G.107.1
[35]). As an alternative method of assessing call quality, R-Factor is scaling from 0 to 120. It
can also be calculated from MOS-LQO value according to ITU-T P.863 – POLQA [27].

Abstract formula

VoLTE Speech Quality R - Factor (received R - party) = f (G.107 - LQ )

VoLTE Speech Quality R - Factor (received H - party) = f (G.107 - LQ )

The KPI shall be tested at the call terminating side and at the call originating side. The two
streams shall be successfully tested at the same time. The table below shows a proposed
mapping range between the user’s satisfaction, MOS and R-Factor scales.

V0.1 Page 64 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

User satisfaction level MOS scale R-Factor scale

Very satisfied 4.3 - 5.0 90 - 100
Satisfied 4.0 - 4.3 80 - 90
Some users satisfied / dissatisfied 3.6 - 4.0 70 - 80
Many users dissatisfied 3.1 - 3.6 60 - 70
Nearly all users dissatisfied 2.6 - 3.1 50 - 60
Not recommended 1.0 - 2.6 < 50
Table 3: Proposed mapping MOS and R-Factor scales

ViLTE Audio Quality (SpQ MOS-LQO) on sample basis

Abstract definition
The KPI of ViLTE audio speech quality on sample basis is an indicator representing the
quantification of the end-to-end voice transmission quality as perceived by the user. This
parameter computes the speech quality on a sample basis. See also 6.7.9 of [2].

Abstract formula

ViLTE speech quality on sample basis (received R - party) = f (MOS - LQO )

ViLTE speech quality on sample basis (received H - party) = f (MOS - LQO )

The KPI shall be tested at the call terminating side and at the call originating side. The two
audio streams shall be successfully tested at the same time together with a successful test
on the two video streams.

Optionally it might be useful to aggregate both speech quality values into one. In this case
the worst of both shall be used. This aggregated speech quality value shall be called SpQ
(min).

ViLTE Video quality (MOS)

ViLTE video quality KPI is characterized by the Mean Opinion Score according to PEVQ
MOS [41].

Abstract definition
End-to-end quality of the video signal as perceived by the end user during a ViLTE session.
This parameter computes the video quality on a sample basis. See also 6.7.10 of [2].

Abstract formula

ViLTE video quality on sample basis (received R - party) = f (PEVQ - MOS)

ViLTE video quality on sample basis (received H - party) = f (PEVQ - MOS)
The KPI shall be tested at the video call terminating side and at the video call originating
side. The two video streams shall be successfully tested at the same time together with a
successful test on the two audio streams.

V0.1 Page 65 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

4.1.3 SRVCC (PS-CS) Quality Parameters [34]

SRVCC is the quality of voice mobility between 4G and the legacy technologies. For a valid
KPI calculation the following preconditions need to be met:

• EPS bearer should be established and the EPS ATTACH message should contain
"IMS PS Voice" as well as "CS Voice" in the voice domain preference list.
SRVCC can have different variants depending on the supporting mid-call, alerting or pre-
alerting state of a VoLTE call when the call is handed over to UMTS or GSM.

SRVCC Success Ratio [34]

Abstract Definition
The KPI is the probability that UE successfully handover a VoLTE call to UMTS or GSM.

Abstract Formula

SRVCC Success Ratio [%] =

 successful SRVCC handovers  100
 all SRVCC invocations
SRVCC Time [34]
For a valid KPI calculation the following preconditions need to be met:

• EPS bearer should be established and the EPS ATTACH message should contain
"IMS PS Voice" as well as "CS Voice" in the voice domain preference list.
• An SRVCC handover to UMTS or GSM is successful.

Abstract Definition
The KPI specifies the time taken to successfully handover a VoLTE call to UMTS or GSM.

Abstract Formula

SRVCC Time [s] = (tSRVCC handover successful ly completed − tSRVCC handover invoked )[s]

4.2 VoLTE / ViLTE Networks & Transport QoS Parameters

The VoLTE / ViLTE networks and transport KPIs contains EPS bearer quality, RTP transport
quality and DIAMETER KPIs.

4.2.1 VoLTE / ViLTE EPS Bearer Quality Parameters

This section contains the EPS bearer KPIs for SIP signalling, voice and video media in the
VoLTE / ViLTE context.

Default EPS Bearer Context Activation Success Ratio

Default EPS bearer context is established and connected to the IMS well-known APN. If a
VoLTE / ViLTE capable UE is involved in the KPI test, then a single or dual default EPS

V0.1 Page 66 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

bearer contexts can be established in the EPS attach procedure. Only the default bearer
context for the IMS well-known APN connection is concerned for this KPI.

Abstract formula

QCI 5 Default EPS Bearer Context Activation Success Ratio [%] =

(To IMS well - known APN)  PDN connection establishment successes
 100
(To IMS well - known APN)  PDN connection initiations
See section 2.2.10 for more details.

Default EPS Bearer Context Activation Time

Default EPS bearer context is established and connected to the IMS well-known APN.

If a VoLTE / ViLTE capable UE is involved in the KPI test; a single or dual default EPS
bearer contexts can be established in the EPS attach procedure. Only the default bearer
context for the IMS well-known APN connection is concerned for this KPI. See also 2.2.11.

Abstract formula

QCI 5 Default EPS Bearer Context Activation Time [s] =

(t IMS PDN connection establishm ent - t IMS PDN connection initiation ) [s]

The default QCI 5 EPS bearer context is used in AM for SIP Signalling and SMSoIP.

Default EPS Bearer Downlink APN AMBR

Each UE in state EMM-REGISTERED is associated with the following bearer aggregate
level QoS parameter: UE Aggregate Maximum Bit Rate (UE-AMBR). The UE-AMBR is
limited by a subscription parameter stored in the HSS. The KPI indicates the APN Aggregate
Maximum Bit Rate by the network in Downlink; the value taken from APN aggregate
maximum bit rate, is an information element assigned by EPC [46].

Default EPS Bearer Uplink APN AMBR

Similar to the downlink APN AMBR, the KPI indicates the APN Aggregate Maximum Bit Rate
by the network in Uplink; the value taken from APN aggregate maximum bit rate, is an
information element assigned by EPC [46].

Default EPS Bearer QCI

The KPI indicates the Quality of Service Class Identifier (QCI) for the default EPS bearer;
the value taken from EPS Quality of Service information element assigned by EPC [46].
QCI=5 is expected.

Dedicated EPS Bearer Context Activation Success Ratio

Dedicated EPS bearer context is established to transport the IMS-based voice media. The
dedicated EPS bearer context for VoLTE is a Persistent EPS bearer context. See also
2.2.12.

V0.1 Page 67 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula

Dedicated EPS Bearer Context Activation success Ratio [%] =

 dedicated EPS bearer activation successes  100
 dedicated EPS bearer activation initiations
The dedicated QCI1 GBR bearer is used in UM for the conversational voice service of
VoLTE and ViLTE audio.

The dedicated QCI2 GBR bearer or QCI 8/9 non-GBR bearer is used in AM for the ViLTE
conversational video service.

Dedicated EPS Bearer Context Activation Time

Dedicated EPS bearer context is established to transport the IMS-based voice media. The
dedicated EPS bearer context for VoLTE is a Persistent EPS bearer context. See also
2.2.13.

Abstract formula

Dedicated EPS Bearer Context Activation Time [s] =

(t Dedicated EPS Bearer context activation success - t Dedicated EPS Bearer context activation initiation ) [s]

The dedicated QCI1 GBR bearer is used in UM for the conversational voice service of
VoLTE and ViLTE audio.

The dedicated QCI2 GBR bearer or QCI 8/9 non-GBR bearer is used in AM for the ViLTE
conversational video service.

Dedicated EPS Bearer Downlink GBR

The KPI indicates the Guaranteed Bit Rate by the network in Downlink; the value taken from
EPS Quality of Service information element assigned by EPC [46].

Dedicated EPS Bearer Uplink GBR

The KPI indicates the Guaranteed Bit Rate by the network in Uplink; the value taken from
EPS Quality of Service information element assigned by EPC [46].

Dedicated EPS Bearer QCI

The KPI indicates the Quality of Service Class Identifier (QCI) for the Dedicated EPS bearer;
he value taken from EPS Quality of Service information element assigned by EPC [46].

V0.1 Page 68 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

IP data volume Rx

Abstract definition
IP data volume Rx denotes the total volume (in Kbytes or Mbytes) of IP data received at Rx
on an EPS bearer. The IP header is included in the calculation.

Abstract formula

Nr
IP data volume received =  IP_PacketLength [KB]
1

Nr is the total number of RTP packets received at Rx.

If QCI5, the KPI measures the IP data volume of SIP signalling and SMSoIP within a
determined period.

If QCI1, the KPI measures the voice IP data volume within a VoLTE or ViLTE call duration.

If QCI2 or QCI 8/9, the KPI measures the video IP data volume within a ViLTE call duration.

IP data volume Tx

Abstract definition
IP data volume Tx denotes the total volume (in KB or MB) of IP data transmitted from Tx on
an EPS bearer. The IP header is included in the calculation.

Abstract formula

Nt
IP data volume transmitted =  IP_PacketLength [KB]
1

Nt is the total number of RTP packets transmitted from Tx.

If QCI5, the KPI measures the IP data volume of SIP signalling and SMSoIP within a
determined period.

If QCI1, the KPI measures the voice IP data volume within a VoLTE or ViLTE call duration.

If QCI2 or QCI 8/9, the KPI measures the video IP data volume within a ViLTE call duration.

4.2.2 VoLTE / ViLTE Media Transport Quality Parameters

The VoLTE media uses symmetric RTP over UDP to transport voice [42]. The RTP quality is
the key factor for the media transport. Because of the symmetry of the RTP protocol [43] the

V0.1 Page 69 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

RTP KPIs in this section are relevant for the receiver Rx and transmitter Tx of the both ends
(call originator and terminator).

The ViLTE media uses two separate RTP connections to transport audio and video streams.
The audio and video RTP quality KPI are separately measured when two RTP connections
are present during the test.

The media transport quality KPIs defined in this section are end-to-end quality parameters. It
is assumed that for a typical VoLTE / ViLTE call between 2 UEs (particularly in an active test
between two probes), the RTP/RTCP is sent end-to-end between the UEs. The RTP packets
can traverse a bunch of RTP-“translators”, e.g. IMS-AGW and TrGW, if network-to-network
interface (NNI) is encountered. The translators do not change the RTP packet payload
contents apart from the Network Address and port translation (NAPT) function (i.e. only IP
header and probably also UDP header are changed).

RTP Max & Mean Packet Delay Variation Rx

The RTP Packet Delay Variation is measured at the e2e Tx/Rx pair on the incoming stream
at Rx [44].

Abstract definition
An RTP packet transit time D is the transfer time of a voice / video RTP stream from Tx to
Rx individually identifiable IP packet, observed at Tx and Rx. D = (R – T) where T is the RTP
timestamp of the packet at Tx, and R is the arrival time of the packet at Rx in the voice /
video RTP stream (of interest within a measurement time interval).
D is also called RTP packet one-way-delay, which is relying on the clock (Time-Stamp)
synchronization between Tx and Rx – a kind of two points time measurement.

The mean RTP packet one-way-delay (OWD) is the average of delay of a voice / video RTP
stream from Tx to Rx. The test accuracy of OWD relies on the clock synchronization
between Tx and Rx.

Taking the minimum transit time in the voice / video RTP stream as the reference, a packet
delay variation PDVi is the difference between the transit time and the reference in the voice
/ video RTP stream . PDVi = Di - Dmin [45], where Di = (Ri – Ti ) is the ith packet transit
time, Ti is the RTP timestamp of packet i at Tx, and Ri is the arrival time of the packet i at Rx
in the voice / video RTP stream within a measurement time interval.

Dmin is the transit time of the packet with the lowest value for delay (minimum) over the
current test interval. Values of PDV may be zero or positive, and quantiles of the PDV
distribution are direct indications of delay variation. PDV is a version of the one-way-delay
distribution, shifted to the origin by normalizing to the minimum delay.

An RTP max PDV is the difference between the maximum transit time and the minimum
transit time in the voice / video RTP stream within a measurement time interval, measured at
Rx.

An RTP mean PDV is the difference between the average transit time and the minimum
transit time in the voice / video RTP stream within a measurement time interval, measured at
Rx.

V0.1 Page 70 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula

RTP max PDV [ms] = ( Dmax − Dmin ) [ms]

 Di
RTP mean PDV [ms] = ( i
− Dmin ) [ms]
nr

 Di
RTP One Way Delay [ms] = i
− (t _ offset _ Rx − t _ offset _ Tx) [ms]
nr

Where nr is the number of received packets in the voice / video RTP stream of interest; Dmin
is the minimum delay of RTP packets in transit time in the voice / video RTP stream of
interest within a measurement time interval.

Assumption
The calculation of the KPI RTP average PDV assumes that the clock offset to the universal
time at Tx (toffset_Tx) and at Rx (toffset_Rx) are stable within the KPI measurement time interval.

RTP means interarrival jitter of incoming stream Rx

The RTP mean inter-arrival jitter is measured at the e2e Tx/Rx pair on the incoming voice /
video RTP streams at Rx.

Abstract definition
The interarrival jitter J is the mean deviation (smoothed absolute value) of the difference in
transit time of two consecutive RTP packets from Tx to Rx in the voice / video RTP stream of
interest as defined in clause 6.4.1 in [42].

Abstract formula

| Di ,i-1 | −J i
J i +1 [ms] = J i + [ms]
16

Where the ith difference Di,i-1 in transit time of two consecutive RTP packets Di,i-1 = (Ri – Ti ) -
(Ri-1 – Ti-1 ), Ti is the RTP timestamp of packet i at Tx, and Ri is the arrival time of packet i at
Rx in the voice / video RTP stream of interest within a measurement time interval.

Assumption
The calculation of the KPI RTP mean that interarrival jitter assumes that the clock offset to
the universal time at Tx (toffset_Tx) and at Rx (toffset_Rx) are stable within the KPI measurement
time interval.

V0.1 Page 71 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

RTP Mean Data Rate Tx / Rx

Abstract definition
This parameter describes an average data transfer rate measured over the entire voice /
video call at Tx / Rx. The voice / video call shall be successfully terminated. The prerequisite
for this parameter is LTE network and IP / UDP service access.

Note: The KPI is measured at RTP level (RTP header + voice / video payload). IP and
UDP headers are not taken into account.

Abstract formula

RTP {Tx | Rx} Data Rate [kbit/s] =

 RTP data transferred [kbit]
(t packet data trans fer complete - t packet data trans fer start )[s]

RTP packets lost Rx

Abstract definition
The KPI is defined as the number of lost packets at Rx side (downlink) in the RTP stream of
interest within the voice / video call duration. This value is equivalent to the number of
missing RTP sequence numbers at Rx.

Abstract formula

RTP_NumPac ketsLost =  lostPackets

RTP packet loss ratio Rx

Abstract definition
The KPI denotes the probability that an RTP packet is lost and not received at Rx.

Abstract formula

RTP_PacketLossRatio [%] =
 RTP_NumPac ketsLost
 100
 RTP_NumPac ketsReceived +  RTP_NumPac ketsLost

RTP_NumPacketsReceived is the total number of RTP packets received at Rx in the RTP

stream of interest within the VoLTE / ViLTE call duration.

V0.1 Page 72 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

RTP round trip delay (RTD)

Abstract definition
An e2e (endpoint A and B) RTP round-trip-delay (A-B-A) is the sum of the packet transit time
Da2b and the packet transit time Db2a of the same RTP packet looped-back. The delay time to
process and loop the packet back at B side is not counted in the RTD.

Abstract formula

RTPmean RTDa 2 b 2 a [ms] =

Dia 2 b +  Dib 2 a
[ms]
N ra

Where RTDa2b2a denotes a packet transit time on a round trip path from endpoint A to B and
from B loopback to A. Dia2b is the transit time of the ith packet from Tx at endpoint A to Rx of
endpoint B. Dib2a is the transit time of the same ith packet looped back from Tx at endpoint B
to Rx of endpoint A. Nra is the total number of looped back packets received at Rx of A side.

Note: Lost RTP packets at looping back are not counted in the calculation.

Assumption
The calculation of the KPI RTP means that RTD assumes the clock offset to the universal
time at Tx (toffset_Tx) and at Rx (toffset_Rx) of the A and B endpoints are stable within the KPI
measurement time interval.

4.2.3 DIAMETER Quality Parameters

Policy & Charging Control

The Diameter PCC KPIs are measured over Rx interface for the quality of Diameter
signaling flows [38] between P-CSCFs and PCRFs. The Rx messages consist of four pairs
of Diameter Request – Answer commands which contribute to a number of PCC procedures
for the initial provision, modification, termination of a Diameter Rx session and session
binding (a Diameter Rx session associated with a VoLTE IMS session) [38].

AA-Request (AAR) Success Ratio

Abstract Definition
The KPI is the probability that an AA-Request command (AAR) sent from P-CSCF via Rx
interface to PCRF results in a successful AA-Answer command (AAA) from PCRF received
by P-CSCF.

Abstract Formula

AAR Success Ratio [%] =

 Number of AAA with Result - Code as Diameter_ Success
 100
 Number of AAR

V0.1 Page 73 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

RA-Request (RAR) Success Ratio

Abstract Definition
The KPI is the probability that a Re-Auth-Request command (RAR) sent from PCRF via Rx
interface to P-CSCF results in a successful RA-Answer command (RAA) from P-CSCF
received by PCRF.

Abstract Formula

RAR Success Ratio [%] =

 Number of RAA with Result - Code as Diameter_ Success
 100
 Number of RAR

ST-Request (STR) Success Ratio

Abstract Definition
The KPI is the probability that a Session-Termination-Request command (STR) sent from P-
CSCF via Rx interface to PCRF results in a successful ST-Answer command (STA) from
PCRF received by P-CSCF.

Abstract Formula

STR Success Ratio [%] =

 Number of STA with Result - Code as Diameter_ Success
 100
 Number of STR

AS-Request (ASR) Success Ratio

Abstract Definition
The KPI is the probability that an Abort-Session-Request command (ASR) sent from PCRF
via Rx interface to P-CSCF results in a successful AS-Answer command (ASA) from P-
CSCF received by PCRF.

Abstract Formula

RAR Success Ratio [%] =

 Number of RAA with Result - Code as Diameter_ Success
 100
 Number of RAR

DIAMETER Routing Agent

The DRA KPI is to measure the DRA routing quality, in order to reach PCRF or HSS (routing
to and redirect / relayed from), when more than one PCRF or HSS has been deployed in a
Diameter realm.

V0.1 Page 74 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Success Ratio of IMS Diameter Messages Routing

The KPI tests the success ratio of all Diameter sessions established over the Cx [39], Sh
[40], Zh interfaces that reach the corresponding HSS when multiple and separately
addressable HSSs have been deployed in a Diameter realm.

• Redirect / route the diameter messages from the HSS over the corresponding
reference points to Cx, Sh, Zh to the corresponding IMS node I/S-CSCF, VoLTE /
ViLTE Application Server, IP Short Message Gateway and Bootstrap Service
Function.

Abstract definition
The KPI denotes the probability that Diameter messages (requests and responses)
transmitted over Cx- ,Sh- , and Zh-interfaces are successfully routed or redirected / relayed
by the DRA. The service failure messages initiated by the DRA over the Cx, Sh, and Zh
interfaces are excluded in the DRA forwarding failures.

Abstract formula

IMS Diameter Routed Messages Success Ratio [%] =

 Requests and Responses successfully transmitted over Cx, Sh, Zh  100
 Total Requests and Responses received by DRA | Cx, Sh, Zh

Success Ratio of PCC Diameter Messages Routing

This KPI measures the success ratio of all Diameter sessions established over the Gx [37],
Rx [38] reference points for a certain IMS session.

• All diameter messages reach the corresponding PCRF when multiple and separately
addressable PCRFs have been deployed in a Diameter realm
• Redirect / route the diameter messages from the PCRF over the corresponding
reference point Gx or Rx to the corresponding IMS node PCEF/PDN-GW or P-
CSCF/IMS-AGW respectively.

Abstract definition
The KPI denotes the probability that of Diameter messages (requests and responses)
transmitted over Gx- and Rx-interface are successfully routed or redirected / routed by the
DRA. The service failure messages initiated by the DRA over the Gx and Rx interfaces are
excluded in the DRA forwarding failures.

V0.1 Page 75 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract formula

PCC Diameter Messages Routed Success Ratio [%] =

 Requests and Responses successfully transmitted | Gx, Rx  100
 Total Requests and Responses received by DRA | Gx, Rx

5 5G NSA (Option 3.X) – Dual Connectivity

This section contains the KPIs related to identify accessibility, connectivity and quality of
network performance allocating Dual Connectivity resources over the New Radio to provide
enhanced mobile broadband over 5G.

• 5G coverage indication
• NR RSRP
• DC Restricted
• 5G ENDC_Used allocation
• Default EPS bearer QoS Parameters
• HTTP Data Access Success ratio
• MBB Download Rate Mbits/s
• MBB Upload Rate Mbits/s
• MBB Latency milliseconds

5.1 5G NSA Coverage Indication

The KPI has the intention to reflect when the UE is capable to detect the New Radio access
over the air and corresponding icon indication of 5G in UE screen. After LTE attach with 5G
capable UE.

Abstract definition
Verification of systemInformationBlock layer 3 message, the presence of
UpperLayerIndication for Release 15 which indicates the UE has entered under coverage
area offering 5G capabilities. Which will be conveted in “Boolean” (true/false) metric
endc_available (enabling dual connectivity)

Abstract formula

∑ 𝑒𝑛𝑑𝑐𝑎𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒 𝑇𝑅𝑈𝐸
5𝐺 𝐶𝑜𝑣𝑒𝑟𝑎𝑔𝑒 𝐼𝑛𝑑𝑖𝑐𝑎𝑡𝑖𝑜𝑛 % =
∑ 𝑇𝑜𝑡𝑎𝑙 𝐴𝑡𝑡𝑒𝑚𝑝𝑡𝑠

V0.1 Page 76 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

5.2 NR RSRP
RSRP (Reference Signal Received Power) the measurement of the power of the main signal
you're getting from a single cell tower (gNodeB)

Abstract definition

For LTE and 5G RSRP is a better indicator of signal strength, which combined with RSRQ
(Reference Signal Received Quality) and SiNR (Signal over noise) will provide an holistic
view of Radio Quality for the target network.

Abstract formula

∑ 𝑁𝑅 𝑅𝑆𝑅𝑃 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑚𝑒𝑛𝑡𝑠 𝑡𝑎𝑘𝑒𝑛

𝑁𝑅 𝑅𝑆𝑅𝑃 [𝑑𝐵𝑚] =
∑ 𝑇𝑜𝑡𝑎𝑙 𝑆𝑎𝑚𝑝𝑙𝑒𝑠

(*) Average metric in dBm

5.3 DC NR Restricted
DC NR Restricted metric is a validation of subscriber profile to be allowed or denied to get
dual connectivity resources from New Radio.

Abstract definition
The metric comes from LTE Attach procedure process where the UE and Network validates
“Use of dual connectivity with NR is not restricted “. By concept the KPI generated based on this will
be “negative” (0% Not restricted/permitted – 100% Fully restricted/not allowed)

Abstract formula

∑ 𝐷𝐶𝑁𝑅_𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑇𝑅𝑈𝐸
𝐷𝐶𝑁𝑅𝑅𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 % =
∑ 𝑇𝑜𝑡𝑎𝑙 𝑆𝑎𝑚𝑝𝑙𝑒𝑠

5.4 5G ENDC_Used allocation

5G Used refers to effective allocation of Dual Connectivity

V0.1 Page 77 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract definition
The KPI will indicate when network effectively allocated the dual connectivity resources for
the UE, which means the user plane will be handled by g-nodeb new radio.

Abstract formula

∑ 𝑒𝑛𝑑𝑐_𝑢𝑠𝑒𝑑 𝑇𝑅𝑈𝐸
5𝐺𝑈𝑠𝑒𝑑 % =
∑ 𝑇𝑜𝑡𝑎𝑙 𝑆𝑎𝑚𝑝𝑙𝑒𝑠

5.5 HTTP Data Access Success ratio

Reference to section 2.9.1 in this document.

The KPI will indicate the IP-Service access to target URL service provider and the success
in establish the http connection with destination.

5.6 MBB Download Rate Mbits/s

Reference to section 2.6.3 in this document.

The KPI defines the data rate taken for download certain amount of Bytes. Method utilized is
based on payload.

The same can be influenced by service-side factors such us radio conditions, cell resource
allocation criteria, network usage, mobile station distance.

5.7 MBB Upload Rate Mbits/s

Reference to section 2.6.3 in this document.

Same definition than 5.6 but in upload direction.

5.8 MBB Latency milliseconds

Reference to section 2.6.5 in this document.

KPI indicates the time taken to send a package and receives echo reply based on ping
method.

Note: For KPIs 5.6, 5.7, 5.8 it is recommended to execute a http speed test and keep
controlled which server is used considering where the UE is receiving the IP access.

V0.1 Page 78 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

6 5G SA (Option 2)

6.1 VoNR – Voice Services

6.1.1 5G Signal to Noise

The signal-to-noise ratio (SINR) refers to the ratio between the power of the desired output
signal and the background noise, which is measured in dB.

Abstract definition
The KPI will indicate the impact of interference into account and is the ratio of signals to
interference and noise.

Abstract formula

𝑆𝐼𝑁𝑅 𝑑𝐵 = 𝑎𝑣𝑒𝑟𝑎𝑔𝑒(𝑆𝐼𝑁𝑅)

6.1.2 VoNR Call Quality Score

5G VoNR Call Quality refers to audio quality perceived by user. It is being calculated based
on POLQA alghoritm.

Abstract definition
The KPI will indicate the audio quality perceived calculated by comparison on reference
audio sent vs audio recorded on other side using POLQA alghoritm which will provide a
score from 1 to 5 (the higher the better)

Abstract formula

𝑀𝑂𝑆 = MOS_LQ

6.1.3 VoNR Call Setup Time

5G VoNR call setup time refers to first establishment of call against the core network.

Abstract definition
The KPI will indicate the network delay to route the call from originator to destination.

Abstract formula

𝐶𝑎𝑙𝑙𝑆𝑒𝑡𝑢𝑝 𝑇𝑖𝑚𝑒 [𝑚𝑖𝑙𝑙𝑖𝑠𝑒𝑐𝑜𝑛𝑑𝑠] = 𝐴𝑉𝐺 (𝐼𝑛𝑣𝑖𝑡𝑒𝐴𝑐𝑐𝑒𝑝𝑡𝑒𝑑 − 𝐶𝑎𝑙𝑙𝐴𝑡𝑡𝑒𝑚𝑝𝑡)

6.1.4 VoNR Call Drop Rate

Call Drop Rate refers to the “abnormal” release of call.

V0.1 Page 79 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Abstract definition
The KPI will indicate when network abnormally release the call interrupting it from the normal
code expected from one of the sides. Reflectes in the SIP session closure with SIPByeEnd.

Abstract formula

∑𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙𝐶𝑎𝑙𝑙𝑅𝑒𝑙𝑒𝑎𝑠𝑒
𝐶𝑎𝑙𝑙 𝐷𝑟𝑜𝑝 𝑅𝑎𝑡𝑒 % =
∑𝐶𝑎𝑙𝑙𝐶𝑜𝑛𝑛𝑒𝑐𝑡

6.1.5 Jitter
Jitter refers to variation in delay between packets in during the voice call.

Abstract definition
The KPI will indicate the difference delay between consecutives packets. This metric is being
calculated in both directions Tx and Rx.

Abstract formula

𝐽𝑖𝑡𝑡𝑒𝑟𝑅𝑋 [𝑚𝑖𝑙𝑙𝑖𝑠𝑒𝑐𝑜𝑛𝑑𝑠] = 𝑅𝑇𝑃_𝐴𝑣𝑔𝐽𝑖𝑡𝑡𝑒𝑟𝑅𝑥

𝐽𝑖𝑡𝑡𝑒𝑟𝑇𝑋 [𝑚𝑖𝑙𝑙𝑖𝑠𝑒𝑐𝑜𝑛𝑑𝑠] = 𝑅𝑇𝑃_𝐴𝑣𝑔𝐽𝑖𝑡𝑡𝑒𝑟𝑇𝑥

6.1.6 Packet Loss

Packet Loss rate refers to amount of data which does not reach destination and it cause
degragadation in the communication.

Abstract definition
The KPI will indicate the percentage of packets lost during the voice call by calculating the
packets which never arrive to destination diving the total packets sent. The metric is in both
side Tx and Rx.

Abstract formula

𝑃𝑎𝑐𝑘𝑒𝑡𝐿𝑜𝑠𝑡𝑇𝑥 ∗ 100
𝑃𝑎𝑐𝑘𝑒𝑡 𝐿𝑜𝑠𝑠 𝑇𝑥 % =
𝑃𝑎𝑐𝑘𝑒𝑡𝑠𝑇𝑥

𝑃𝑎𝑐𝑘𝑒𝑡𝐿𝑜𝑠𝑡𝑅𝑥 ∗ 100
𝑃𝑎𝑐𝑘𝑒𝑡 𝐿𝑜𝑠𝑠 𝑅𝑥 % =
𝑃𝑎𝑐𝑘𝑒𝑡𝑠𝑅𝑥

6.1.7 Latency
Latency refers to the time audio signal takes to travel from source to destination.
Abstract definition

V0.1 Page 80 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

The KPI will indicate the average delay between the reference and the test signal.
Abstract formula

𝐿𝑎𝑡𝑒𝑛𝑐𝑦 [𝑚𝑖𝑙𝑙𝑖𝑠𝑒𝑐𝑜𝑛𝑑𝑠] = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒_𝐷𝑒𝑙𝑎𝑦

6.1.8 Data Rate

Data Rate or (BitRate) refers to incoming and outgoing stream from each side

Abstract definition
The KPI will indicate in kbits per second the incoming Rx streams and outgoing Tx streams
from each side during the communication.

Abstract formula

𝑘𝑏𝑖𝑡
𝐵𝑖𝑡𝑅𝑎𝑡𝑒 𝑇𝑥 [ ] = 𝑎𝑣𝑔(𝑅𝑇𝑃𝐷𝑎𝑡𝑎𝑟𝑎𝑡𝑒𝑇𝑥 )
𝑠

𝑘𝑏𝑖𝑡
𝐵𝑖𝑡𝑅𝑎𝑡𝑒 𝑅𝑥 [ ] = 𝑎𝑣𝑔(𝑅𝑇𝑃𝐷𝑎𝑡𝑎𝑟𝑎𝑡𝑒𝑅𝑥 )
𝑠

6.1.9 5QI
5G Quality of service identifier

Abstract definition
The KPI will indicate the QoS id assigned after attach procedure. The same can be modified
later on when specific service request is demanded by the UE. Would be possible to get
“default 5QI” and once service is selected “dedicated 5QI”

Abstract formula

5𝑄𝐼 = 5𝑄𝐼

6.1.10 SST
Service Slice Type

Abstract definition
The KPI will indicate which Slice Type was allocated by network following the 3gpp standard

Slice/Service type SST Characteristics

value
eMBB 1 Slice suitable for the handling of 5G enhanced
Mobile Broadband.
URLLC 2 Slice suitable for the handling of ultra- reliable low
latency communications.
MIoT 3 Slice suitable for the handling of massive IoT.
V2X 4 Slice suitable for the handling of V2X services.

V0.1 Page 81 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

HMTC 5 Slice suitable for the handling of High-Performance

Machine-Type Communications.

Abstract formula

𝑆𝑆𝑇 = 𝑆𝑆𝑇

6.2 SMS over 5G

6.2.1 SMS Submission Success Rate

SMS was delivered to the SMSC and originating part received the confirmation.

Abstract definition
The KPI will indicate the “OK” on sending the sms from originating part.

Abstract formula

𝑅𝑒𝑐𝑒𝑖𝑣𝑒𝑑_𝐴𝐶𝐾 ∗ 100
𝑆𝑀𝑆 𝑆𝑢𝑏𝑚𝑖𝑡 % =
𝑇𝑜𝑡𝑎𝑙 𝐴𝑡𝑡𝑒𝑚𝑝𝑡𝑠

6.2.2 SMS Submission Delay

SMS submission time to the SMSC.

Abstract definition
The KPI will indicate the time in milliseconds on sending the sms from originating part and
getting the ACK from SMSC that was effectively submited.

Abstract formula

𝑆𝑀𝑆 𝑆𝑢𝑏𝑚𝑖𝑡 𝐷𝑒𝑙𝑎𝑦 = [SMS_SubmitAck − SMS_Submit]

6.2.3 SMS End To End Delivery Success Rate

SMS was delivered to the Destination party.

Abstract definition
The KPI will indicate the “OK” on SMS Delivery to the destination party.

Abstract formula

V0.1 Page 82 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 < 𝑏 − 𝑠𝑖𝑑𝑒 >∗ 100

𝑆𝑀𝑆 𝐸2𝐸 𝐷𝑒𝑙𝑖𝑣𝑒𝑟𝑦 % =
𝑇𝑜𝑡𝑎𝑙 𝐴𝑡𝑡𝑒𝑚𝑝𝑡𝑠

6.2.4 SMS End To End Latency

Time taken for SMS to be delivered to the destination party.

Abstract definition
The KPI will indicate the time in milliseconds taken for the sms delivery from origination party
to the destination party.

Abstract formula

𝑆𝑀𝑆 𝐿𝑎𝑡𝑒𝑛𝑐𝑦 = [SMS_Received − SMS_Submit

7 Single Service Indicator and Single Quality Indicator

7.1 Single Service Indicator

Single service indicator (SSI) is a unique quality performance indicator which composes a
set of selected KPIs of a particular service. Each KPI in the set has a weight according to the
relevance contributing to the single service indicator.

7.1.1 Purpose of SSI

The main purpose of using SSI is to provide a single unique QoS score for the executive
management, instead of many individual KPIs.

7.1.2 SSI Calculation

Calculation of an SSI consists of two steps, KPI normalisation and weighting.

KPI normalisation
Depending on the KPI definition, different KPIs can have different units, for example,
percentage (%) or data rate kbit/s, etc. KPI normalisation maps the KPI measured value into
a range between 0 and 100, by using two scoring limits, a higher limit and a lower limit for
each KPI (higher limit > lower limit). After the KPI value normalisation, the higher the KPI
score is, the better the quality is implied.

Despite of divergence of KPI definitions, there are two types of KPIs which have different
calculations before the normalisation.

V0.1 Page 83 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

14. Higher is better, for example, success ratio, data rate

15. Lower is better, for example, loss ratio, latency delay or session establish time
For scoring "higher is better":
16. KPI value less than the lower limit is interpreted as 0
17. KPI value greater than the higher limit is interpreted as 100
18. KPI value between the two limits is linear interpolated using formula:

KPI value - lower limit

normalised value =  100
higher limit - lower limit

For scoring "lower is better":

19. KPI value greater than the higher limit is interpreted as 0
20. KPI value less than the lower limit is interpreted as 100
21. KPI value between the two limits is linear interpolated using formula:

higher limit - KPI value

normalised value =  100
higher limit - lower limit

KPI weighting
The weight of a KPI defines how important the role of this particular KPI is in the contribution
of the SSI calculation. Assigning a bigger or heavier weight to a KPI will give the KPI more
influence on SSI. Assigning smaller or lighter weights to KPIs will make those KPIs not so
important. Negative weights are not allowed in the SSI calculation, but using zero weight can
temporarily disable the KPI without removing it from the SSI definition. There has been at
least one non-zero weighted KPI in the SSI definition.

7.1.2.1 SSI calculation formula

The following formula is used for the SSI calculation:

 normalised KPIi  weighti

SSI = 1
n


1
weighti (all weights)

where "normalised KPIi" is a normalised value of the ith particular KPIi and "weighti" is the
value of this KPIi's "scoring weight". A total n KPIs are included in the SSI calculation.

Some examples helps for better understanding:

22. If all KPIs contained in a SSI have the same weight, the value of SSI will be simply
arithmetic average of all KPIs. It means all KPIs play the same role in the calculation
of the SSI.
23. If the weight of a KPI is set to 0, this particular KPI is ignored or masked at the SSI
calculation.
24. An SSI with two KPIs weighted with 1 and 3, will be calculated using 25% of the first
KPI's value and 75% of the second KPI.

V0.1 Page 84 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

7.2 Single Quality Indicator

Similar to SSI combining the weighted KPI and resulting in a single quality score value for a
particular service, single quality indicator (SQI) composes weighted SSIs of different
services and provides a unique single score value for the quality.

7.2.1 SQI calculation

A SQI value can be calculated by using the following formula

 SSIj  weightj
SQI = m
1


1
weightj (all weights)

where "SSIj" is a value of the jth particular SSIj and "weigh tj" is the value of this SSIj's
"scoring weight". A total of m services, i.e. SSIs, are included in the SQI calculation.

Annex A Examples for measuring trigger points

A.1 SMS-Service:

A.1.1 Layer 3 Messages:

Start SMS Service Attempt: Generating random access (chan_request
SDCCH) at mobile equipment

Successful SMS Service Receiving cp_data (rp_ack) at mobile equipment

Attempt

Receiving SMS on Mobile Receiving cp_data (rp_ack) at mobile equipment

Equipment 2:

V0.1 Page 85 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

V0.1 Page 86 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Annex B Document Management

B.1 Document History

Approval Editor /
Version Date Brief Description of Change
Authority Company
0.1.0 5/1/2001 First draft of document for IREG
QoS WP discussion
0.2.0 30/3/2001 Second draft of document for
IREG QoS WP e-mail discussion
0.3.0 27/4/2001 Third draft for IREG QoS
workshop including Data Services
1.0.0 10/6/2001 First stable Version for chapters
Telephony, SMS
2.0.0 11/9/2001 Document for approval at IREG
#41 and GSMA with a document
classification of “Unrestricted –
Public”
3.0.0 21/9/2001 Document approved by GSMA

3.0.1 30/4/2002 Parameter Completion Rate SMS

circuit switched added
3.0.2 14/5/2002 Parameter for CSD and PSD Data
Services added
3.1.0 17/6/2002 Document for approval at IREG
#43 and GSMA with a document
classification of "Unrestricted –
Public"
3.2.0 10/10/2002 Changes in chapters 2.4, 2.5-2.7
approved by QoSWP Meeting #8,
02.10.2002
3.2.1 10/2/2003 Editorial changes in chapters 2.5-
2.7
3.3 16/4/2007 IREG doc 52_037 incorporated
“Addition of the new QoS
parameter definitions”
3.4 10/8/2009 Signal doc 43_009 incorporated
“Changing computation method Marko Onikki,
for speech quality from MOS to TeliaSonera
PESQ ”
3.5 9/12/2009 Incorporate outputs of Global IREG#55, Marko Onikki,
roaming Quality project EMC#79 TeliaSonera
3.6 4/1/2011 Signal Docs 49_16 and 50_009 Marko Onikki,
IREG Signal
incorporated TeliaSonera
4.0 6/6/2011 Signal Docs 53_009 incorporated Marko Onikki,
IREG Signal
TeliaSonera
5.0 22/5/2015 Incorporated CR1001 and Marko Onikki
IREG
CR1002 TeliaSonera
6.0 5/11/2015 Incorporated CR1003 Marko Onikki
NG
TeliaSonera
7.0 18/10/2016 CR1004 Add VoLTE and ViLTE Marko Onikki
NG
roaming quality parameters Telia Company
8.0 29/8/2017 CR1005 (Introduction of SSI and Marko Onikki
NG
SQI calculations Telia Company
9.0 7/6/2018 CR1006 and CR1007 Marko Onikki
NG
Telia Company

V0.1 Page 87 of 88
GSM Association Confidential - Full, Rapporteur, and Associate Members
Official Document IR.42 Definition of Quality of Service parameters and their computation

Approval Editor /
Version Date Brief Description of Change
Authority Company
10 14/06/2024 CR1008 - IR81Alignment Javier
NG-ISAG Sendin/GSMA

11 2/01/2025 CR1009 IR.42 - 5G SA Quality of Manuela

Service parameters NG-ISAG Montagna,
Hutchinson

B.2 Other Information

Type Description
Document Owner GSMA NG
Editor / Company Marko Onikki / Telia Company

Feedback

It is our intention to provide a quality product for your use. If you find any errors or omissions,
please contact us with your comments. You may notify us at prd@gsma.com

Your comments or suggestions & questions are always welcome.

V0.1 Page 88 of 88

Definition of Quality of Service Parameters and Their Computation 14 June 2024
No ratings yet
Definition of Quality of Service Parameters and Their Computation 14 June 2024
87 pages
Definition of Quality of Service Parameters and Their Computation
No ratings yet
Definition of Quality of Service Parameters and Their Computation
39 pages
Definition of Quality of Service Parameters and Their Computation 4.0 6 June 2011
No ratings yet
Definition of Quality of Service Parameters and Their Computation 4.0 6 June 2011
37 pages
Definition of Quality of Service Parameters and Their Computation 17 October 2016
No ratings yet
Definition of Quality of Service Parameters and Their Computation 17 October 2016
75 pages
Definition of Quality of Service Parameters and Their Computation 29 August 2017
No ratings yet
Definition of Quality of Service Parameters and Their Computation 29 August 2017
76 pages
IR.42 v9.0
No ratings yet
IR.42 v9.0
77 pages
Doc2-Radio Networks QoS Assessment and Evaluation Tools
No ratings yet
Doc2-Radio Networks QoS Assessment and Evaluation Tools
46 pages
Service Performance Report
No ratings yet
Service Performance Report
21 pages
Kpi2 PDF
No ratings yet
Kpi2 PDF
140 pages
SOP For TSP - Updated
No ratings yet
SOP For TSP - Updated
26 pages
QoS Compliance in GSM & 3G Networks
No ratings yet
QoS Compliance in GSM & 3G Networks
17 pages
Etsi TS 102 250-1
No ratings yet
Etsi TS 102 250-1
12 pages
MUZ To ISB RD
No ratings yet
MUZ To ISB RD
23 pages
2018 - Chapter 2 - KPI of Network Performance and Quality of Service - Sigit Haryadi
No ratings yet
2018 - Chapter 2 - KPI of Network Performance and Quality of Service - Sigit Haryadi
14 pages
MCI Iran QoS Low Level Design
No ratings yet
MCI Iran QoS Low Level Design
20 pages
UMTS Capacity Estimation
No ratings yet
UMTS Capacity Estimation
70 pages
LTE Network Performance KPIs - Validation
No ratings yet
LTE Network Performance KPIs - Validation
25 pages
Measurements and QoS Analysis SwissQual
No ratings yet
Measurements and QoS Analysis SwissQual
6 pages
Performance Enhancement of GSM System: Qos Analysis and Optimization
No ratings yet
Performance Enhancement of GSM System: Qos Analysis and Optimization
4 pages
05 WO - NP2003 - E01 - 0 UMTS Capacity Estimation-33
No ratings yet
05 WO - NP2003 - E01 - 0 UMTS Capacity Estimation-33
34 pages
p1746 Pitas PDF
No ratings yet
p1746 Pitas PDF
5 pages
Verifying Network Performance and KPIs
No ratings yet
Verifying Network Performance and KPIs
52 pages
Bharti Quality Workshop-3G - KPI-Kolkata
100% (1)
Bharti Quality Workshop-3G - KPI-Kolkata
32 pages
QoS Requirements For VoIP
No ratings yet
QoS Requirements For VoIP
95 pages
ELK 1302 191 Manuscript 1
No ratings yet
ELK 1302 191 Manuscript 1
25 pages
09-KPI Training
100% (1)
09-KPI Training
13 pages
Huawei-WCDMA Capacity Planning
No ratings yet
Huawei-WCDMA Capacity Planning
86 pages
Performance Enhancement Optimization of GSM System Through The Analysis of Key Performance Indicator For QoS
No ratings yet
Performance Enhancement Optimization of GSM System Through The Analysis of Key Performance Indicator For QoS
5 pages
Etsi Eg 202 934
No ratings yet
Etsi Eg 202 934
28 pages
Tata Network Performance Nov 2010
No ratings yet
Tata Network Performance Nov 2010
18 pages
LTEPMCounter20150911140851 KPI
No ratings yet
LTEPMCounter20150911140851 KPI
3,057 pages
Category Name: eRAN13.1 KPI Reference
No ratings yet
Category Name: eRAN13.1 KPI Reference
24 pages
KPI OSS (BSC Level - Busy Hour)
No ratings yet
KPI OSS (BSC Level - Busy Hour)
2 pages
LN1040 LN0093 38GHz 56M 422 Rollout
No ratings yet
LN1040 LN0093 38GHz 56M 422 Rollout
32 pages
2G KPI Acceptance Target and Methodology AoP 2010
No ratings yet
2G KPI Acceptance Target and Methodology AoP 2010
14 pages
Service Excellence. Delivered: Gprs+Umts+Lte Intro
No ratings yet
Service Excellence. Delivered: Gprs+Umts+Lte Intro
45 pages
Regulations TRAI QoS Report
No ratings yet
Regulations TRAI QoS Report
15 pages
GSM Service KPI
No ratings yet
GSM Service KPI
67 pages
3G TOPS Formula
No ratings yet
3G TOPS Formula
27 pages
WO - NAST3018 - E01 - 1 UMTS QoS Management 50P
No ratings yet
WO - NAST3018 - E01 - 1 UMTS QoS Management 50P
50 pages
Huawei ERAN KPI Reference Summary v1r15c00
100% (1)
Huawei ERAN KPI Reference Summary v1r15c00
24 pages
S11P1 Samuel-Agyekum
No ratings yet
S11P1 Samuel-Agyekum
15 pages
Methodology
No ratings yet
Methodology
11 pages
Tobiet-Lorenz2000 Chapter PerformanceMeasurementMethodol
No ratings yet
Tobiet-Lorenz2000 Chapter PerformanceMeasurementMethodol
16 pages
Call Setup Success Rate in GSM Network Using Citizen Sensing Approach: A Proposal
No ratings yet
Call Setup Success Rate in GSM Network Using Citizen Sensing Approach: A Proposal
5 pages
Assignment 1: Advance Wireless Networks
No ratings yet
Assignment 1: Advance Wireless Networks
4 pages
Quality of Service KPIs
No ratings yet
Quality of Service KPIs
3 pages
Telecom KPI Reference Guide
No ratings yet
Telecom KPI Reference Guide
28 pages
Volte Network Architecture Overview: Ims Core
No ratings yet
Volte Network Architecture Overview: Ims Core
23 pages
3G Network Performance KPIs
0% (1)
3G Network Performance KPIs
16 pages
Back-To-Back User Agent (B2bua) Sip Servers Powering Next Generation Networks
No ratings yet
Back-To-Back User Agent (B2bua) Sip Servers Powering Next Generation Networks
31 pages
NGN Architectures Protocols and Services
No ratings yet
NGN Architectures Protocols and Services
2 pages
VoLTE: Architecture and Bearer Setup
No ratings yet
VoLTE: Architecture and Bearer Setup
6 pages
VOLTE
100% (1)
VOLTE
26 pages
MAPS™ IMS Test Suite
No ratings yet
MAPS™ IMS Test Suite
8 pages
1final Iot MODULE 4
No ratings yet
1final Iot MODULE 4
56 pages
Blade
No ratings yet
Blade
4 pages
Telecom Network Specialist Profile
No ratings yet
Telecom Network Specialist Profile
3 pages
CSC3300 V100R009C01 Feature Description
No ratings yet
CSC3300 V100R009C01 Feature Description
307 pages
5G SA Option 2 ImplementationGuideline v1.2
100% (2)
5G SA Option 2 ImplementationGuideline v1.2
42 pages
CFX 5000v10
100% (2)
CFX 5000v10
14 pages
Cv-Anglaishager Ben Fradj
No ratings yet
Cv-Anglaishager Ben Fradj
5 pages
Triple Play Service Delivery Architecture
No ratings yet
Triple Play Service Delivery Architecture
12 pages
CV Ashwani Gautam
No ratings yet
CV Ashwani Gautam
4 pages
ADDAX Solutions GD Dvs2
100% (1)
ADDAX Solutions GD Dvs2
28 pages
Clearwater
No ratings yet
Clearwater
187 pages
iCG9815 V300R003C00 Product Description
No ratings yet
iCG9815 V300R003C00 Product Description
48 pages
OCS Design Document Overview
No ratings yet
OCS Design Document Overview
13 pages
1 - Introduction To TAS
No ratings yet
1 - Introduction To TAS
32 pages
Volte Call Flow
No ratings yet
Volte Call Flow
10 pages
Introduction To IPTV, Technology, Standards & Systems
No ratings yet
Introduction To IPTV, Technology, Standards & Systems
36 pages
AbrarShah (11 0)
No ratings yet
AbrarShah (11 0)
4 pages
PE'S - CC5T - Cohort 07
No ratings yet
PE'S - CC5T - Cohort 07
12 pages
3gwireless Secured Itlab - Com.vn
No ratings yet
3gwireless Secured Itlab - Com.vn
5 pages
HUAWEI AR2200 Series Enterprise Routers Datasheet PDF
No ratings yet
HUAWEI AR2200 Series Enterprise Routers Datasheet PDF
20 pages
Huawei IPTV Solution Technical Proposal Template For Designer Based On The Version V4.0
100% (1)
Huawei IPTV Solution Technical Proposal Template For Designer Based On The Version V4.0
87 pages
TSV-1000-8S FXS VoIP Gateway Details
No ratings yet
TSV-1000-8S FXS VoIP Gateway Details
4 pages
Shari Rogers Resume
No ratings yet
Shari Rogers Resume
7 pages
IMS Resume
No ratings yet
IMS Resume
2 pages
3GPP TS 23.207
No ratings yet
3GPP TS 23.207
46 pages

Definition of Quality of Service Parameters and Their Computation 27 January 2025

Uploaded by

Definition of Quality of Service Parameters and Their Computation 27 January 2025

Uploaded by

GSMA

Definition of Quality of Service parameters and their

Security Classification: Non-Confidential

2.4.20 CSFB Return to LTE Time 34

2.9.2 FTP {Download|Upload} IP Service Access Setup Time [s] [2] 48

4.1.2 VoLTE / ViLTE Service Quality Parameters 59

7.2.1 SQI calculation 85

1.1 Scope of document

Figure 1: Relation of IR.42 to other PRDs in GRQ framework

The framework consists of four PRDs and a trouble report database.

1.2 General considerations

• the service is available and not barred for any reason;

1.3 Definition of Terms

POLQA Perceptual Objective Listening Quality Analysis

1.4 Document Cross-References

No. Document Description

No. Document Description

2.1 QoS Parameters Model

Figure 2: Four Layer model for QoS Parameters.

2.2 Service-independent QoS parameters

2.2.1 Network Availability [2] [12]

GPRS: broadcasting GPRS indicator in system information 3 or system information 4

UMTS (WCDMA) [25], LTE [26]: S Criteria fulfilled

When the mobile equipment supports multi-mode (GSM/UMTS/LTE/5G),the judgement on

For CSFB and SMSoSGs, LTE S criteria is applied.

probing attempts with mobile services available

2.2.2 Network Accessibility [2]

NA-CS[%] is applied to GSM or UMTS-CS. NA-PS[%] is applied to GPRS, UMTS-PS, LTE,

2.2.3 Circuit Switched LU Success Ratio (CS LU – SR) [2]

See ETSI 102 250-1 (5.2 Network non-accessibility).

Stop: Mobile receives the CS attach accept message.

2.2.4 Circuit Switched Location Update Delay (CS LU - D)

Sum(t CSAttachComplete − t CSAttachStart )

Start: Point of time when the mobile sends the attach

2.2.5 Packet Switched LU Success Ratio (PS LU - SR)1 [2] [7]

See ETSI TS 102 250-2 (5.3 Attach Failure Ratio)

Stop: Mobile receives the PS attach accept message.

A combined EPS + IMSI attach is applied to CSFB or SMSoSGs.

2.3 Packet Switched Location Update Delay (PS LU - D) [2]

See ETSI TS 102 250-2 (5.4 Attach Setup Time)

Sum(t PSAttachComplete − t PSAttachStart )

Stop: Point of time when the mobile receives the attach

Any PDP Context Activation request is considered as an attempt.

LTE: see Default EPS Bearer Context Activation Success Ratio

2.3.2 PDP Context Activation Time [2] [19]

See ETSI TS 102 250-2 (5.6 PDP Context Activation Time)

End: Reception of the notification of successful PDP

User activation (GPRS PDP context creation procedure) is recommended as network

2.3.3 PDP Context Cut-Off Ratio [2] [8]

ETSI TS 102 250-2 (5.7 PDP Context Cut-off Ratio).

See also 3GPP TS 29.060.

A precondition for measuring this parameter is that a PDP context is successfully

Additional Default EPS Bearer Context Activation Success Ratio [%] =

Any additional Default Context Activation request is considered as an initiation.

2.3.5 Default EPS Bearer Context Activation Time [2]

Additional default EPS bearer context:

Additional Default EPS Bearer Context Activation Time [s] =

2.3.6 Dedicated EPS Bearer Context Activation Success Ratio [2]

Dedicated EPS Bearer Context Activation success Ratio [%] =

2.3.7 Dedicated EPS Bearer Context Activation Time [2]

Dedicated EPS Bearer Context Activation Time [s] =

EPS Bearer Context Cut-off Ratio [%] [2]

Default | Dedicated EPS Bearer Context Cut - off Ratio [%] =

2.3.8 DNS Host Name Resolution Success Ratio [%] [2]

DNS Host Name Resolution Success Ratio [%] =

The KPI is relevant only for PS services.

2.3.9 DNS Host Name Resolution Time [s] [2]

The KPI is relevant only for PS services.

2.4 Telephony Service

A precondition for the applicability of this section to the voice CSFB:

2.4.1 Service Accessibility Telephony - MO (SA-T-MO) [2] [12] [14]

See complementary Service Non-Accessibility Telephony (SNAT) in ETSI TS 102 250-2

Successful call attempt: Measurement (alerting4 tone heard or ALERTING

H-party rings or sends ALERTING (CC message).

2.4.2 Service Accessibility Telephony - MT (SA-T-MT) [2] [14]

See complementary Service Non-Accessibility Telephony (SNAT) in ETSI TS 102 250-2