 Now, we are going to look at another very interesting application. In fact, an increasingly important application in NGN, i.e., IP television. We'll start with the introduction, then we'll look at and touch upon some interesting aspects including the coding requirements in terms of data rate, the limitations, and then we'll look at the IP television industry as an end-to-end domain. So, ITUT actually has defined a television standard for IP-based networks to offer a variety of services including to but not limited to the broadcast television as such, video content, audio content, pictures, graphics, and data. So all this heterogeneous kind of data comes under the umbrella of IPTV. So, the ITUT standard actually recommends certain support services that need to be provided. For instance, quality of service, quality of experience, interactivity, that is control in the hands of the user. It has to be reliable service and of course, certain encryption-based security mechanism. So IPTV basically is the most natural successor to the broadcast television networks. In fact, before IP television came in, the analog broadcasting had already been replaced with digital TV broadcasting over both terrestrial and cable networks but bringing it to IP actually requires a whole new set of services. So, digital television as compared to analog is based on data. So, it needs a higher bit rate compared to other services because there's video content and on top of it is superimposed audio in terms of audio channels. So, there's a requirement to identify some minimum data rate available on the X side and at the core side. For that, broadband access networks are the most natural contenders to offer this service. The coding and coding requirements are such that for standard definition television, the terminology standard definition television is best understood vis-a-vis high definition television called HDTV but for now we'll keep the scope limited to standard definition television. So, SDTV actually requires a data rate of 2 to 3 megabits per second. Now in a typical terrestrial broadcast system or even cable, we have tuning capability. So users actually tune in to different channels by switching to that particular frequency but in IPTV this is not going to be the case so users actually are not required to tune in. In fact, users subscribe to a certain IPTV stream and for that there is a whole mechanism for multicasting. The broadcasting is not affordable in IP based networks so multicasting is carried out at the core and the transport networks that is over long haul networks and then on the X side it is usually the unicasting mechanism which is implemented. Unicasting actually helps need to do basis delivery that is whenever a user is interested in using the video content of a certain service provider then only then that particular stream is unicast to that particular user. And this is done on the basis of channel to channel whichever channel the user is currently interested in viewing so it is not going to be all the time available service so when a user in an IP based network switches from one IPTV channel one to the second IPTV on channel two that essentially means stopping the stream of the channel one and initiating the stream of channel two so but this again is limited because we have a broadband connection which is shared by telephony services by non IPTV based interactive applications and the best effort traffic. So if you look at the transmission mechanism in IPTV the basic design and the basic delivery mechanism is such that the data it becomes the upper bound or the limiting factor. So there is a limitation to the maximum number of simultaneous channels which can be delivered from the network side to the user equipment and of course the limiting factor is mainly the bandwidth and that to the bandwidth of the access side because on the core and the transport side the bandwidth provisioning is less of a problem and it all already is implementing multicasting but when it comes to the to an individual user or a home subscriber then bandwidth on the access side is a limitation so in order to handle of course like we switch on a typical television remote a channel to and then we move on to another channel and to another channel this actually can be accommodated through its equivalent implementation through admission control. So it means that if you look at typical DSL kind of connectivity ADSL for instance we have eight megabits per second on the downlink. So we can accommodate say two channels or up to maximum four channels if there's no other service which is running but usually two is recommended on the downlink and the uplink is used for transmission of certain control information for instance you want to change the channel from one to the other. So with this IPTV mechanics are quite clear. Now we need to look at the overall umbrella of IPTV which actually requires certain domains to be identified and each domain needs special consideration investment design challenges and of course a lot of engineering opportunity. So we have the end user site that is a home subscriber we have the network provider which is providing the networking infrastructure that is NGN to deliver the content then we have the service provider service provider is the one that is offering us certain services if that means the channels that we are subscribing to are coming from the service provider and then we have the content provider this content provider is the one that is creating the content generating the videos and the live and the stream content for instance movies news sports etc. And now this particular content provider may have a many to one relationship with the service provider or may have a one to many relationship where we have multiple content providers for a single service provider and a single content providers in turn connected to multiple service providers. This is a very interesting way of looking at how the overall domains of IPTV are worth considering.