Contents

Contents............................................................................................................................. 1 Chapter 1............................................................................................................................ 2 Introduction........................................................................................................................ 2 1.1 Problem Statement....................................................................................................... 2 1.2 Objectives..................................................................................................................... 3 1.3 Significance .................................................................................................................. 3 1.4 Methodology................................................................................................................. 3 Chapter 2............................................................................................................................ 4 Literature review................................................................................................................. 4 2.1 Digital Television Solutions............................................................................................5 2.1.1 Web-based Video....................................................................................................... 5 2.1.2 Digital Video Broadcast..............................................................................................5 2.2 Next Generation Networks and the IMS.........................................................................6 2.3 IP Television.................................................................................................................. 8 2.3.1 Dedicated IPTV Systems.............................................................................................9 2.3.2 IMS-Based IPTV.......................................................................................................... 9 2.4 Direct Marketing Platform IMS-based IPTV...................................................................10 Conclusion........................................................................................................................ 16 2.5 The challenges of voice-over-IP-over-wireless.............................................................17 2.5.1 Service requirements for voice.................................................................................19 Unequal error detection (UED)..........................................................................................19 Unequal error protection...................................................................................................19 End-to-end delay...............................................................................................................20 Conclusion........................................................................................................................ 20

Chapter 1 Introduction
The Internet explosion has made the flow of information more powerful and the access to content ubiquitous. The value of the Internet lies in its ability to connect directly to consumers when and where they want. This provides content and service providers the ability to reach consumers across multiple platforms. The introduction of telecommunication services over the Internet has increased the demand for reliable communication services. IP-based telecommunication networks have been designed to provide rich value-added multimedia communication services to subscribers. These services, which are offered over highly managed Service Delivery Platforms (SDPs), offer guaranteed Quality of Service (QoS) and Quality of Experience (QoE) levels. The IP Multimedia Subsystem (IMS) is a Next Generation Network (NGN) control framework which is designed to provide reliable communication services and multimedia applications to subscribers, by providing the control functionality required to run managed SDPs over NGN networks. The IMS is able to provide critical enhancements to service delivery when compared to Internet-based services. These include QoS and QoE guarantees, convergence of services and billing, as well as personalisation of services. The IMS is an IP-based framework that enables convergence and personalisation of services. For this reason, it has been chosen as the platform for the VoD frame work which is proposed in this project work.

1.1 Problem Statement

As revenues from voice services continue to fall, telecommunication operators have begun to develop and acquire new services in an attempt to generate new revenue streams and to avoid losing customers to third party service providers. One such service

that many, if not all, operators are investigating into, is Multimedia services using Internet Protocol (IP). This is the motivating factor for which video on demand (VoD) over IMS was proposed.

1.2 Objectives
This project proposes to design and implement VoD using IMS platform

1.3 Significance
In our modern day of technological advancement, subscribers must be able to access videos on demand. With the help of computers, Pads, androids, etc., connected to a network or service provider, subscribers must be in a position to watch variety of videos at their convenience. These videos may include tutorials, music videos, movies, documentaries, etc,. We believe that with such a network designed, clients will be in a position have control over what they watch.

1.4 Methodology
Information regarding to the success of this project will be gathered from all relevant sources: books, documentations and the internet with references to all. With the design, we will be building an Application Server (AS) running Linux Operating System software. The End user will be running IMS client software that will grant access to the network. We will also create a Server that will be housing the videos.

Chapter 2 Literature review

Over the past ten years, there has been an increase in subscriptions for three digital services: digital television, high speed data and wireless services. Figure 2.1 illustrates this growth in the United States of Americas domestic market, where an increase in demand for all three digital services can clearly be seen. Figure 2.1: USA domestic growth of digital services Broadband subscriptions increased from 50 million in 2005 to 75 million two years later in 2007, and is expected to reach 100 million by 2015. A similar growth rate can be seen for digital television.

Figure 2.1: USA domestic growth of digital services Broadband subscriptions increased from 50 million in 2005 to 75 million two years later in 2007, and are expected to reach 100 million by 2015. A similar growth rate can be seen for digital television. This chapter investigates digital Television solutions. It also examines the different types of IPTV, NGN and IMS in general. Lastly, it looks at the implementation of delivering personalised Advertisement in IMS-based IPTV through direct marketing as well as the challenges of Voice-Over-IP.

2.1 Digital Television Solutions

Digital television solutions can be divided into Web-based video and Digital Video Broadcast (DVB). Both types offer access to digital video media, although their approaches differ.

2.1.1 Web-based Video

Web-based video is IP based and may be accessed through any PC with an Internet connection. As mentioned earlier, this limits QoS due to the best effort nature of the open Internet. Furthermore, the users QoE is limited due to the size and resolution of the PC screen and the low quality video that is generally delivered over the Internet. Broadcast television is not supported by web-based solutions. Content is either user generated or on demand. These services do not generally require a subscription. YouTube and iPlayer are examples of Web-based video implementations.

2.1.2 Digital Video Broadcast

Three types of DVB services exist; these are Satellite DVB (DVB-S), Cable DVB (DVBC) and Terrestrial DVB (DVB-T). DVB services may be accessed through a STB connected to a television and either a satellite, cable or terrestrial connection. Both broadcast and on demand services are supported by DVB solutions. Since these services are offered over a closed and managed platform, unlike Web-based video, they do not suffer from limited QoS and QoE levels, but they do require a subscription. Implementations of Web-based video have also moved to television sets, requiring a STB, an Internet connection and a subscription. This is more of a hybrid solution, merging web-based video and DVB, e.g., WebTV. Such implementations support broadcast and on demand video and can be classified as IPTV solutions. They are standalone services, falling into the category of dedicated IPTV systems. These systems

are generally bundled with broadband services, allowing some level of convergence. However, for complete service convergence, IMS-based IPTV is required.

2.2 Next Generation Networks and the IMS

The IP Multimedia Subsystem (IMS) is an architectural framework which provides the necessary control functionality required to run managed SDPs over NGN networks. AllIP networks enabled by the IMS and SDPs offer possibilities for reliable and secure communication services and multimedia applications for subscribers. The IMS was originally standardised by the 3GPP as a service control layer to run over their IP-based 3G networks. The converged IP core in NGN networks, which forms the control plane, constitutes the IMS. For this reason, the control plane is also known as the IMS layer. NGN networks are heterogeneous, allowing access to the IMS layer through various access networks. This enables support for both fixed and mobile users. Its architecture is divided into four planes: the application plane, control plane, transport plane and access plane. Figure 2.2 shows a high level schematic of the NGN architecture. The IMS core is made up of SIP servers called Call Session Control Functions (CSCFs) that are responsible for session control. There are three types of CSCFs, namely the Proxy CSCF (P-CSCF), the Interrogating CSCF (I-CSCF) and the Serving CSCF (SCSCF). The P-CSCF handles the control functions of the SIP-based communication and is the first point of contact between the client and the IMS network. . It acts as a SIP proxy server and forwards SIP requests and responses in the appropriate direction. It also performs security related functions. The I-CSCF retrieves user location information and routes the SIP requests to the appropriate destination, typically to an S-CSCF.

Figure 2.2: High level diagram of the NGN architecture The S-CSCF is the central node of the control plane. The entire SIP signalling in the IMS traverses the allocated S-CSCF. It inspects every SIP message and determines whether the SIP signalling should visit one or more Application Servers (AS) on its way towards its final destination. The Home Subscriber Server (HSS), also located in the control plane, contains all the user related subscription information required to handle multimedia sessions. This information includes location information, security information, user profile information and the SCSCF which is allocated to the user. The application plane contains SIP Application Servers that host and execute services. Each

AS may operate in one of four modes: SIP proxy mode, SIP User Agent (UA) mode, SIP redirect mode, or SIP Back-to-Back User Agent (B2BUA) mode, to connect two SIP User Agents. The media plane and charging domain host media and charging functions respectively. Media functions are responsible for delivering media streams to the client and allowing trick play functions. Media streams do not traverse the IMS core; media delivery takes place directly between the client and the media plane. However, it is the responsibility of the IMS core elements to facilitate the delivery of the media stream according to the available network resources and client terminal capabilities. The charging functions allow for both online and offline charging. The relevant AS creates charging events and sends these to the charging domain where the appropriate credit checks or updates are done. If online charging is active, credit checks are done when a user requests a service from the AS. With offline charging, credits are updated after a user ends the session. Since NGN networks are converged and run highly managed SDPs, they are able to offer QoS, QoE and security guarantees, using the control functionality of the IMS. They also allow for convergence of services and billing, personalisation of services and integration with legacy networks. Furthermore, due to its layered architecture, rapid service creation is possible. Offering services over NGN networks, using IMS functionality, allows these services to reuse well standardised IMS components to enforce IP control.

2.3 IP Television
There are three types of IPTV services. These are broadcast/live television services; on demand/unicast services (VoD); and PVR services which allow recording, pause and time shifting of the video stream. Many different IPTV architectures have been proposed by the various standardization bodies and are generally vendor specific. However, they can all be classified as one of two types of architectures: dedicated IPTV or IMS-based IPTV. Although these are two different approaches, they both have the

common goal of delivering managed TV and video to the user over an IP channel to enhance the traditional television experience. Enhancements include recording capabilities, time shifting and a personalised service. IPTV is able to combine the flexibility of Web-based video with the reliability of DVB and WebTV to deliver a personalised and highly managed digital TV service to the subscriber.

2.3.1 Dedicated IPTV Systems

Dedicated IPTV Systems use a dedicated subsystem within an NGN platform to provide all the required IPTV functionality, such as service control and user profile management, to subscribers. These systems have the advantage of dedicated resources. However, they are closed proprietary solutions. As a result, inter-working with other NGN elements to provide a converged service becomes difficult. As a result, dedicated IPTV systems are generally used as standalone services.

2.3.2 IMS-Based IPTV

IMS-based IPTV allows the IPTV service to make use of embedded IMS functionality such as authentication, authorisation and accounting. This ensures minimal wastage of resources in the network and ensures convergence, since a common IMS core is used to provide the control functionality for all NGN services. Standard bodies such as the Open IPTV Forum (OIPF) are pushing IMS/NGN based telecommunication services for convergence. Other advantages of offering IPTV over the IMS are support for single sign on, subscription and session management, roaming, service personalisation and unified charging and billing. The IMS-based IPTV architecture standardised by the European Telecommunications Standards Institute (ETSI) Telecoms and Internet Converged Services and Protocols for Advanced Networking (TISPAN) is the most widely accepted IMS-based IPTV architecture and most comprehensive set of specifications. They are developed in

cooperation with other standardisation bodies such as ITU-T IPTV Focus Group, ATIS IPTV Interoperability Forum and DVB, to ensure interoperability between vendor specific IPTV systems. For this reason, it is the most widely adopted architecture in evolving standards and research. Figure 2.3 illustrates the architecture of the ETSI TISPAN IPTV system, including the relevant entities, reference points and protocols used in a typical IPTV session. The User Equipment (UE) is a functional entity that provides the user with access to IPTV services. It is the users point of contact with the IPTV service. The Core IMS contains the CSCFs, whose functions were explained in the previous section. The User Profile Selection Function (UPSF) is equivalent to the HSS in 3GPP specifications; it is a database that stores user profiles and any IPTV specific profile data. The Media Control Function (MCF) is a functional entity that provides the UE with functions required to control media flows. It also manages the Media Delivery Functions (MDFs) which are under its control. The MDF is a functional entity that delivers media content to the UE.

2.4 Direct Marketing Platform IMS-based IPTV

Telecommunication companies (Telcos) all over the world are rolling out the next generation networks (NGN) a number of operators already have an Internet protocol multimedia subsystems (IMS) infrastructure in place. Deploying IMS architecture is cost intensive; hence, telcos need means to recover the venture cost within particular timeframes. Direct marketing is a technique that enables product sellers to reach a specific target consumer group which results in a higher turnover. This paper explored the design of an advertisement system for personalized direct marketing to motivate IMS deployment using the University of Cape Town (UCT)

Figure 2.3: ETSI TISPAN IMS-based IPTV architecture

Advanced IP Television (IPTV) server, deployed in the UCT IMS platform. The design is based on the IMS (IPTV) application. IPTV is a system whereby digital television content is delivered to subscribers via an IP network infrastructure. It differs from WebTV in that it is a highly managed system, offering guaranteed quality of service (QoS) and involving user subscriptions. The IMS-based IPTV provides a great avenue for a direct marketing system due to its personalised and interactive nature.

To achieve such an effective interactive and personalised application certain architectural requirements were necessary: 1. Two way communication between the client and the server. IMS uses the Session Initiation Protocol (SIP) to provide this feature. 2. The ability of users to personalised services through modification of user profiles stored in the Home Subscriber Server (HSS) which contains each user attributes used to personalised the user experience. The Fig. 2.4 below show the architecture designed for the purpose, it uses the ESTI TISPAN IMS-based IPTV architecture modified with an advertisement application server (AS). When the IPTV AS receives the forwarded service request message from the S-CSCF, it communicates with the HSS to authorise usage of the service. It then selects the MCF and MDF. But, instead of forwarding its response back to the S-CSCF as in the IPTV implementation, the Advertisement Application Server processes the SIP message further. It queries the HSS to retrieve the appropriate user profile information. With this information it runs an algorithm to categorise the user. Each category is then mapped to a group of advertisement IDs (Ad IDs) using a look-up table. Once these processes are complete, the request is sent back to the S-CSCF with the Ad IDs included in the optional attributes section of the SDP (Session Description Protocol) body in the SIP message. When the message arrives at the MCF, the Ad IDs are used to amend the requested IPTV media stream to include the relevant advertisements selected by the Advertisement AS in the requested media stream. The algorithm running on the Advertisement AS is a basic decision loop. Advertisements are matched to certain categories or combinations of categories before hand, by the marketer, and each advertisement is assigned an ID. These IDs are entered into the look-up table, stored in the Advertisement AS, under their associated category. Once a decision is made by the

Advertisement AS to categorise the user, the relevant Ad IDs, retrieved from the lookup table, are added to the SDP body of the SIP message going to the S-CSCF. This information is extracted by the Media Server on receiving the SIP message. It is then used to control the adaptation of the media stream to include the selected, personalised advertisements. The Algorithmic State Machine (ASM) diagram in Fig.2.5 describes its operation. The figure shows a simplified version that considers only two variables. As the number of variables increases, the complexity of the algorithm will grow exponentially. The first decision is based on gender, thus two possible branches follow. These branches then categorise the user according to age and the loop exits. The resulting category has a gender and age group associated with it.

Fig. 2.4 Advertisemet Platform for ims-based IPTV These categories are predefined and entered in a look-up table. If no information is shared by the user, the algorithm will assign a generic category before exiting the loop.

The personalised advertisement algorithm uses a timer to schedule additional advertisements that would be sent to the UE during the lifetime of a TV session. Upon receipt of the channel request message at the IPTV and Advertisement AS, the timer is started. At the expiry of a timeout period (e.g., 10 minutes), the advertisement algorithm retrieves another Ad ID for that category of user. A SIP (advertisement) message is then sent to the UE, containing the RTSP address of the advertisement. This is a unicast message, sent to the UE. At the UE, the advertisement would appear as a banner across the bottom of the screen with text describing the product or service and prompts the user to click on an RTSP link to view the video advertisement. In this way, the advertisements are non-invasive, hence not degrading the users Quality of Experience (QoE). If the user clicks on the link, the IPTV content is paused and the advertisement is streamed to the user directly from the Media Server. If the user does not click on the link, the advertisement expires after 30 seconds and the banner is removed from the screen. The IPTV AS is notified when the user clicks on the link in order to implement discounts or charging changes, depending on the advertised product. Typical signalling call flows for an IPTV system with and without advertisements are shown in Fig. 2.6and Fig. 2.7 respectively. To validate and evaluate the testbed, three scenarios were studied:
1.

No advertisements are clicked on after the user makes a request to view content. Then after 30 seconds and the user is able to view the requested media.

2. Advertisements are viewed but without any changes to the user profile. 3. Advertisements are viewed but with change(s) to the user profile.

Fig. Decision algorithm run by the advertisement AS

Fig. 2.6 UCT Advanced IPTV call flow

Fig. 2.7 Advertisement framework in UCT IPTV call flow The Challenges encountered during the process are latencies observed during call setup. Further works proposed by this paper will be to improving the framework by offering additional features. Secondly would be to base the algorithm not only on the users age and gender, but also on attributes such as marital status, income bracket and interests.

Conclusion
To conclude, Telcos and third party service providers have solution to their inability to recoup their investments in rolling out IMS-based services. Network operators and service providers have been provisioning marketing and advertisement services in order to make additional profits.

2.5 The challenges of voice-over-IP-over-wireless

Today, the accumulated volume of data traffic is on the verge of surpassing the accumulated volume of voice traffic in all public networks. Given the growth in the areas of wireless voice and data, we see that the combination of mobile and Internet communication constitutes the driving force behind third-generation wireless systems, which promise to support at least 144 kbit/s (384 kbit/s) in all radio environments, and up to 2 Mbit/s in low-mobility and indoor environments. The standardization of thirdgeneration wireless systems is rapidly progressing in all major regions of the world. The main application for 3G services will be wireless packet transfer; for instance, for wireless access to the Internet. However, support will also be provided for high datarate circuit-switched services, such as real-time video. The paper discusses briefly, VoIP service. The basic components of the voice service are two user terminals with IP-based voice applications and a network that provides end-to-end transport between the terminals. The terminals exchange voice samples using the real-time transport protocol (RTP), which has been standardized by the IETF. Also the terminals can establish and maintain communication without the involvement of a third-party entity. However, sometimes the two user endpoints cannot establish end-to-end communication without outside intervention; for example, when they do not know one another Os IP address or do not use the same voice codec. In these cases, a control plane framework is used to route incoming traffic and to negotiate terminal capabilities (codec support, multiparty conferencing, and so on), in traditional telecommunications this functionality, which is referred to as call control, is provided by, say, a GSM mobile switching centre. In the IP world, there are two main methods of providing call control functionality: ITU-T Recommendation H.323 and the IETF session initiation protocol (SIP) .Originally intended for LAN environments, H.323 is an ITU standard for multimedia applications. Today, however, the standard is being adopted for broader usage. H.323 encompasses a complete architecture and a set of

protocols, such as H.225 for call control and H.245 for bearer control. H.323 uses IETF protocols, such as the real-time protocol and the resource reservation protocol (RSVP). Besides end-user terminals, the H.323 architecture encompasses gatekeepers, gateways, and multiparty units. In this context, emphasis is put on the gatekeepers and gateways which constitute the VoIP server. The gatekeeper part is the controlling unit that provides call-control functionality; the gateway part contains the user plane functions. H.323 call control is based on Q.931, which is also used in GSM and ISDN. The session initiation protocol, which is an IETF standard draft, is only one component in the IETF alternative to the H.323 paradigm for a complete multimedia architecture. Other necessary protocols and components include the session description protocol (SDP), the services access point (SAP), and the real-time control protocol (RTCP). The session initiation and session description protocols (SIP/SDP) do not make up architecture; they were designed for session initiation. In contrast to H.323 and GSM/ISDN, SIP/SDP does not provide a complete call-control mechanism; an SIP proxy primarily provides routing and addressing services; device management is not included. However, the SIP proxy (or VoIP server) can be enhanced to include functionality for offering other services such as transcoding. The session initiation protocol is associated with a paradigm in which call control is distributed over several entities, and in which the user terminal plays a central role in coordinating these entities. In summary, the two IP-based terminals exchange voice samples that have been encapsulated in RTP over the IP network. The terminals exchange control signalling between themselves or, with assistance from network entities such as a VoIP server, establish and maintain communication sessions through the network according to either the H.323 or SIP paradigm. Both the session initiation protocol and H.323 support endto-end solutions in which the network solely functions as a bearer. In this case we assume that an SIP or H.323 network call agent can, if so requested, support the terminal with transcoding services. The mobile terminal supports cellular access

(UMTS/WCDMA or EDGE) and a complete VoIP application that is based on either SIP or H.323. It was assumed that an adaptive multirate (AMR) codec will be supported by future VoIP clients. In addition to basic UMTS packet switched access, the network contains functions for adapting media, routing calls, and for authenticating users and services.

2.5.1 Service requirements for voice

Unequal error detection (UED) Ordinarily, the bits in a frame from a cellular voice codec are divided into three classes: 1a, 1b and 2. Bit-error sensitivity varies between these classes Class 1a includes the most sensitive bits; Class 2, the least sensitive bits. In a typical second-generation system, the bits in Class 1a are covered by a cyclic redundancy code (CRC) that checks for errors in the frame. Thus we say that the voice frame uses UED scheme. If information on the different classes of bit-error sensitivity cannot be transferred from the codec to the radio-access network, or if the bits in the voice frame are not organized into classes, then the UED scheme cannot be used. Instead, an equal error detection (EED) scheme a CRC that covers the entire voice frame is introduced. To have the quality in these two cases, each must receive the same number of frames with a bad CRC. Where circuit-switched traffic is concerned, only frames with a bad CRC contribute to the frame error rate. But in an IP based radio network, frames with a bad CRC, frames that are lost due to jitter, and fatal errors in the IP header contribute to the frame error rate. By fatal errors in the IP header we mean user datagram protocol (UDP) checksum errors, errors in the link layer, and header decompression errors.

Unequal error protection The calculation of the bit error rate (BER) only includes those errors that occur in bits not protected by the CRC. The residual errors in bits protected by the CRC must be as

close to zero as possible. If any residual bit errors exist in the Class 1a bits, the voice decoder might produce noticeable artefacts. If UEP is not available (but UED is), the lowest FER requirement of Class 1a and the BER requirement of Class 1b constitute the channel requirements.

End-to-end delay
The ITU-T recommends the following limits on one-way delay: 1. 0-150 ms acceptable for most applications. 2. 150-400 ms acceptable provided the administrator knows what impact the transmission time will have on the transmission quality of other user applications.
3.

Any delay greater than 400 ms unacceptable for general network planning; however, in some exceptional cases (such as satellite hops) this limit will be exceeded.

A typical second-generation cellular system has a one-way delay of less than 100 ms. Thus, to achieve the same high end-to-end quality, the one-way delay requirement for the conversational service (terminal-to- PSTN) of a third-generation wireless system must also be less than 100 ms.

Conclusion
The widespread growth of the Internet has created a mass market for multimedia and information services. The challenge of providing these services via third-generation wireless systems is twofold: from the market perspective, the challenge is to merge the installed base of users in cellular and Internet environments; and in terms of technology, the challenge is to find common denominators for cellular solutions and efficient Internet access. To succeed in meeting these challenges, third-generation wireless systems must be designed to provide a multitude of services, offering considerable

flexibility and cost-effective access with structured quality-of-service handling and ensuring high radio-spectrum efficiency. A key objective of third-generation all-IP networks is to provide a capable service platform for IP-based applications. The solutions we have described for audio streams and associated control protocols will advance the UMTS network another step toward becoming a full-fledged service platform that can support demanding services, such as IP-based conversational multimedia.