Voice over LTE

VoLTE

Miikka Poikselkä, Harri Holma, Jukka Hongisto, Juha Kallio, Antti Toskala

• Publisher: Wiley
• Year: 2012
• Print ISBN: 9781119951681
• eBook ISBN: 9781119945765
• Edition: 1
• Topic: Technology & Engineering
• Subtopic: Mobile & Wireless Communications
• Index: Technology & Engineering

Describes the technological solutions and standards which will enable the migration of voice and SMS services over to LTE/EPC networks

Main drivers for the introduction of Long Term Evolution of UTRAN (LTE) is to provide far better end user experience for mobile broadband services. However, service providers also need to have a clear strategy of how to offer voice and messaging services for consumers and enterprises. The voice service over LTE is becoming increasingly important when the smartphone penetration is increasing rapidly. Smartphones require both good quality voice and high speed broadband data.

This book provides the exhaustive view to industry-approved technologies and standards behind the Voice over LTE (VoLTE). Whether a decision maker or technology analyst, this book explains a topic of substantial global market interest.   It provides a good introduction to the technology and is useful for operators who may be deploying VoLTE, product managers responsible for VoLTE products and those who work in implementation and standardization of related technologies.

• Provides a comprehensive overview of industry-approved technologies and standards, providing vital information for decision makers and those working on the technology
• Written by authors working at the cutting edge of mobile communications technology today, bringing a mix of standards and product background, guaranteeing in-depth practical and standards information
• Covering the technical and practical elements of VoLTE, explaining the various approaches for providing voice services over LTE

Chapter 1

Background

At the end of 2004 the third generation partnership project (3GPP) Standardisation Forum started to evaluate a new radio technology as a successor for wideband code division multiple access (WCDMA). This work was called long term evolution (LTE) and is nowadays the radio interface name used in most official publications. Inside 3GPP the newly developed radio access network is called the evolved UMTS terrestrial radio access network (E-UTRAN) to indicate the path from the global system for mobile communications (GSM)/Enhanced data rates for global evolution (EDGE) radio access network (GERAN) via the GSM/general packet radio service (GPRS)/EDGE to UTRAN [WCDMA/high-speed packet access (HSPA)] and finally to E-UTRAN (LTE). In parallel to the work on a new radio interface 3GPP initiated a study to evolve the 2G/3G packet core network (known as the GPRS core) in order to cope with the new demands of LTE. This core network study was called system architecture evolution (SAE) and it was documented in the 3GPP technical report (3GPP TR 23.882). The final outcome of this work was a new packet core design in Release 8 documented in (3GPP TS 23.401) and (3GPP TS 23.402), called the evolved packet core (EPC). 3GPP Release 8 was officially completed in March 2009 and the world's first commercial LTE network was opened in December 2009 by TeliaSonera.

3GPP Release 8 introduces major advances in mobile networks. For the subscriber, it means higher access rates and lower latency on the connection, while for the mobile communication service provider, LTE radio technology provides lower cost per transmitted bit thanks to more efficient use of radio network resources and delivers excellent voice spectral efficiency, as described in Chapter 7. The technology also offers more flexibility in frequency allocation, thanks to the ability to operate LTE networks across a very wide spectrum of frequencies. LTE also minimises the power consumption of terminals that are used ‘always on’. 3GPP Release 8 also introduces major advances in the core network that improve service quality and networking efficiency, leading to a better end user experience. GPRS technology has already introduced the always on concept for subscriber connectivity and 3GPP Release 8 mandates this ability, with at least one default bearer being always available for all subscribers. This allows fast access to services as well as network initiated services such as terminating voice calls and push e-mail. The connection setup time for person to person communication is also minimised with always on bearers.

But considering the fact that LTE is an all-internet protocol (IP) technology we can get to the conclusion that the voice service will have to be delivered in a different way as circuit switched voice will not be possible. So there is a need for a voice solution on top of LTE. Voice in this IP world, would be implemented as voice over Internet protocol (VoIP). The 3GPP-specified way to support VoIP is the IP multimedia subsystem (IMS). It is an access-independent and standard-based IP connectivity and service control architecture that enables various types of multimedia services to end users using common Internet-based protocols. 3GPP has worked with the IMS since 2000 and there exist thousands of pages, in different specifications, that cover IMS related functionalities. In the meantime, a sophisticated architecture and feature set has been developed. Moreover, 3GPP has specified multiple, different ways to complete single functions (e.g. authentication, session setup, supplementary service execution, bearer setup) which increases complexity of the IMS.

While 3GPP has specified all of the ‘ingredients’ needed to implement IMS-based voice over long term evolution (VoLTE)—such as session initiation protocol (SIP) registration, signalling compression, call set up and supplementary services—it has left it up to communication service providers and vendors to decide which of the numerous alternative implementation options to use. This is frankly a recipe for a chaotic and fragmented rollout of IMS-based VoLTE since there is no way to guarantee that different industry players will opt for the same ‘ingredients’ that their competitors' choose for their own implementations. It goes without saying it was not a model for success.

Unsurprisingly, in the absence of a clear-cut approach to VoLTE, alternatives emerged, most notably 3GPP specified circuit switched fallback (CSFB), in which an communication service provider uses its legacy 2G/3G network to handle voice calls. In this scenario, when an LTE terminal initiates a voice call or receives one from the legacy circuit-switched network, it downgrades any ongoing LTE data session to 3G or HSPA speeds for the duration of the voice call. If the voice call ‘falls back’ to a 2G network, the LTE data session will likely be suspended altogether, as 2G data speeds are not sufficient for broadband data applications. In either case, the impact on customer experience can be obvious.

Another emerged alternative for IMS-based VoLTE was VoLTE via generic access (VoLGA) promoted by the VOLGA Forum.

All major network and handset vendors compete aggressively for the biggest possible slice of network communication service providers' business. But at the same time, their business is an interconnected business, where equipment interoperability, especially between handset and network is the key to ensuring that they can all play together. For communication service providers voice have been the killer application and it is going to be big source of revenue for years to come. When there were three different voice solutions (IMS-based VoIP, CSFB, VoLGA) there was rightfully serious concerns whether LTE would come with voice anytime soon. So the situation in year 2009 was equally challenging for both communication service providers and vendors. This is why, from time to time, serious rivals come together to agree on technical cooperation that is designed to help smooth the way forward for the common good of the telecoms market. On 4 November 2009 the One Voice initiative was published by AT&T, Orange, Telefonica, TeliaSonera, Verizon, Vodafone, Alcatel-Lucent, Ericsson, Nokia Siemens Networks, Nokia, Samsung and Sony Ericsson. These 12 companies announced that they have concluded that the IMS-based solution, as defined by 3GPP, is the most applicable approach to meet their consumers' expectations for service quality, reliability and availability when moving from existing circuit switched telephony services to IP-based LTE services. The companies in One Voice then set about to create a solid foundation for securing the smooth introduction of standards-based VoLTE. They evaluated the different alternative ‘ingredients’ specified by 3GPP in order to settle on a minimum set of essential handset and network functionalities and features that communication service providers would need to implement basic, interoperable VoLTE service. These agreed mandatory set of functionalities for the user equipment (UE), the LTE access network, the EPC network and the IMS functionalities are contained in the ‘technical profile’ published by One Voice, available for use by anyone in the industry. In a sense, the technical profile gives all industry stakeholders a level playing field on which to enhance their VoLTE service as they see fit, but most importantly a level playing field that enables the basic working, and interworking, of VoLTE across the entire industry landscape.

15 February 2010 marks the second important milestone in VoLTE ecosystem development. On that date, the Global System for Mobile Association (GSMA) announced it has adopted the work of the One Voice initiative to drive the global mobile industry towards a standard way of delivering voice and messaging services for LTE (GSMA) and Next Generation Mobile Networks alliance (NGMN) delivered communication service providers' agreement to ensure roaming for VoLTE by recommending to support CSFB in all LTE voice devices and networks (NGMN). The GSMA's VoLTE initiative was supported by more than 40 organisations from across the mobile ecosystem, including many of the world's leading mobile communication service providers, handset manufacturers and equipment vendors, all of whom support the principle of a single, IMS-based voice solution for next-generation mobile broadband networks. This announcement was also supported by 3GPP, NGMN and the International Multimedia Teleconferencing Consortium (IMTC). Following the announcement, work progressed very quickly; and already in March 2010 the GSMA permanent reference document (IR.92) on IMS profile for voice and short message service (SMS) was published containing an improved version of the One Voice profile. In September 2010 GSMA agreed to freeze the content of the permanent reference document (IR.92). A global baseline for commercial VoLTE deployments was finally stabilised.

Chapter 2

VoLTE Deployment Strategies

Global mobile networks have been driven by voice service since the beginning and voice has been the ‘killer’ application for many years. It can be said that there would be no mobile networks without voice services. Today, we have a new situation where data traffic is heavily growing in mobile networks and all deployments are going to be driven by data increase. Also from a standards point of view a data-driven approach has been selected where future mobile networks are based on the internet protocol (IP) technology and mobile services are built on top of the ‘data layer’.

The third generation partnership project (3GPP) long term evolution (LTE) is the main technology for further mobile networks and it can be said already now that it is the fastest developing mobile system technology ever. One reason is that LTE first deployments are for data with dongles and data cards including short message service (SMS) support, but services like voice will follow later. From a LTE deployment perspective the LTE voice service can be regarded as just one data application, but with specific requirements for real-time traffic, quality of service and interoperation with existing voice infrastructure [circuit switched (CS) core networks]. It is also self-evident that voice remains as a mandatory service in LTE.

The LTE/evolved packet core (EPC) standards have been designed as mobile technology based purely on the IP. The LTE/EPC networks will be driven by mobile broadband services in which the voice and SMS traffic amounts are a smaller portion of the total traffic amount but still retain their importance for both subscribers and operator business. From a technology point of view the LTE/EPC network is designed to carry all applications and it has excellent support for voice service with low latency and high capacity, which means savings because in the end the operator does not have to maintain parallel networks for voice and data.

The voice service over LTE is done with an IP multimedia subsystem (IMS) as specified by 3GPP. LTE radio access does not support direct connectivity to CS core and services, but radio is connected to EPC that provides IP connectivity for the end user services and interworking towards existing CS networks.

2.1 Common Networks Everywhere

It is fair to assume a long migration from CS voice to a full blown IMS-based voice over internet protocol (VoIP) and therefore 3GPP has standardised several building blocks to enable different roadmaps towards an all-IP network. With these building blocks a communication service provider can have the right voice introduction based on its own voice strategy, that is more CS reuse or more IMS-driven.

The LTE is a common technology for 3GPP and 3GPP2 operators and from a voice deployment strategy point of view it is good to understand the differences between existing global system for mobile communication (GSM), wideband code division multiple access (WCDMA) and code division multiple access (CDMA) deployments. Both GSM and WCDMA technologies provide simultaneous voice and data service, even though GSM dual transfer mode (DTM) is not as widely used as WCDMA multi radio access bearer (RAB). The CDMA technology is widely limited to evolution data only (EV-DO) where simultaneous voice and data is not possible.

2.2 GSM/WCDMA View

The expected deployment strategy for existing GSM/WCDMA communication service providers is to start with circuit switched fallback (CSFB) and gradually move towards IMS-based VoIP deployments. Notice that either of the interim all-IP steps can be omitted in this evolution. When LTE radio network is introduced the main steps to provide voice service are as follows:

1. CSFB as a first voice solution for LTE subscribers.

2. IMS-based VoIP with single radio voice call continuity (SR-VCC) prior countrywide LTE coverage exists. SR-VCC enables handover from IMS-based VoIP to CS speech when user equipment (UE) is running out of VoIP coverage, and reverse SR-VCC enables handover from CS speech to IMS-based VoIP.

3. IMS-based VoIP where all the calls are made over packet switched networks (all-IP).

In addition to the steps listed above there can also be mixed environments where some voice services use CSFB but others use IMS-based VoIP. For example IMS-based VoIP for a communication service provider's own subscribers and CSFB for inbound roamers.1

2.3 CDMA View

The starting point for an existing CDMA operator is that the voice service is offered over a CS network. When introducing LTE radio network the potential steps to provide voice service are as follows:

1. Simultaneous voice and long term evolution (SV-LTE) or simultaneous voice and data only (SV-DO) where dual radio terminal offers simultaneous LTE/evolved high rate packet data (eHRPD) data and CDMA 1×voice. This allows simultaneous voice and data service use for CDMA users.

2. CSFB is also an option for some CDMA operators, but is not necessary needed if mobiles offer SV-LTE.

3. Eventual target IMS controlled voice over all-IP network where all the calls are made over packet switched networks, either on the LTE only or on the LTE and eHRPD.

Single radio voice call continuity is also standardised from LTE to single carrier radio transmission technology (1×RTT) but there exists the same limitation as on CSFB, that is a simultaneous data and voice service cannot be offered after UE is moved to 1×RTT during a voice call. In general, it can be said that both CSFB and SR-VCC are possible but they require such investment to legacy CDMA technology that the operators' interest seems to be rather full LTE coverage as soon as possible with IMS based VoIP. The intermediate phase will passed mostly with SV-LTE terminals, and CSFB will be used only by some operators.

 

 

1 The CSFB will be used as the initial voice roaming solution because IMS based VoIP roaming is still being finalised in GSMA and 3GPP at the time of writing.

Chapter 3

VoLTE System Architecture

3.1 Overview

This chapter introduces the reader to the voice over long term evolution (VoLTE) system architecture. This includes various network entities, how they interact with each other via different reference points. In this boo...

Table of contents

• Cover
• Title Page
• Copyright
• Preface
• Acknowledgements
• List of Abbreviations
• Chapter 1: Background
• Chapter 2: VoLTE Deployment Strategies
• Chapter 3: VoLTE System Architecture
• Chapter 4: VoLTE Functionality
• Chapter 5: VoLTE End to End and Signalling
• Chapter 6: IMS Centralized Services
• Chapter 7: VoLTE Radio Performance
• Chapter 8: HSPA Voice over IP
• References
• Index