NGN Functional Architecture

The NGN architecture as defined by the ITU and ETSI borrows heavily from the work done by 3GPP. The diagram below shows the architecture that has been defined for the NGN in ITU-T Rec. Y.2012.

NGN Architecture

According to Y.2012, the NGN architecture supports the delivery of multimedia services and content delivery services, including video streaming and broadcasting. An aim of the NGN is to serve as an PSTN and ISDN replacement.

The NGN architecture defines a Network-Network Interface (NNI), User-Network Interface (UNI), and an Application Network Interface (ANI). The Transport stratum provides IP connectivity services to NGN users under the control of Transport control functions, including the Network Attachment Control Functions (NACF) and Resource and Admission Control Functions (RACF).

The NACF and RACF are two components of the NGN. A more detailed component view of the NGN is shown in the following figure:

NGN Components

Terminals that talk to the NGN will authenticate with the Network Attachment Control Functions (NACF), receive an IP address, get configuration information, etc. Once attached to the network, terminals will communicate directly or indirectly with the Resource and Admission Control Functions (RACF) in order to get desired QoS for communication, get permission to access certain resources, etc.

There are a number of components within this architecture which would suggest a "freedom of services" to users. Users would potentially have access to RTSP-based streaming services, PSTN access, IMS access, and access to "other multimedia components". In this author's opinion, the very fact that these components are listed as the components of the NGN suggests that the service providers behind this effort wish to have too much control.

Note that "legacy" terminals (user devices) must pass through a gateway device. Perhaps this makes sense for legacy PSTN equipment, but what about newer IP devices? Are all IP devices, by default, NGN devices? They are not, according to those trying to define the NGN. So, what are IP devices we use today? Are they exclused from the NGN or locked into tight control with limited access? It would seem so, suggesting that the NGN is at odds with the Internet and a threat to the Internet as we know it.

The IMS layer, while certainly suitable for wireless networks, is very much a restrictive component within any kind of "next generation" fixed network. Below is a diagram that shows the functional architecture of the NGN with interfaces to the various functional elements.

IMS Functional Architecture Model

The primary purpose of IMS is to provide control over telephone calls. In a "next generation" network, why would it be that the access provider would, in any way, try to specify how the phone system internals should function? According to ITU-T Rec. Y.2001, access and service should be separated and IMS really has no business being a core part of the NGN. You can look at IMS as nothing more than a means of maintaining the current legacy telephone company business model of counting minutes and charging for minutes of use. That kind of business model is dated and needs to be replaced with something more forward-looking.

The only interesting aspect of the NGN architecture as proposed by the ITU and ETSI is the "other multimedia" components. Through these components, it might be possible to truly deliver a new kind of network that users may find useful. The NASS and RACS layers are quite reasonable, as access to the network must necessarily be controlled. However, once network access is granted, the kinds and types of services accessible to the user should be virtually unlimited (though that does not equate to "free", understand). Through the "other multimedia" components, it might be possible for terminal equipment manufacturers to provide equipment that might deliver the services that users expect today, as well as new kinds of services that users might want.

In fact, the ITU has initiated work on a new multimedia system called H.325, which promises to deliver on the promises of the NGN. Specifically, H.325 will allow a user to use multiple devices togther with multiple applications in order to communicate with another person. As the system is envisaged, any application developer could create an application that simply "plug in" to the H.325 terminal. The user could, for example, use a mobile phone to talk on a phone, see a video feed on an LCD screen, collaborate on a document through a PC ("screen sharing" or "application sharing"), and receive a file to a mobile handheld device, all within the context of a call to another person. The possibilities are amazing.

A very important distinction between the work on H.325 and IMS is that H.325 is designed to work on the Internet as it is today. Look at H.325 as a big step forward in terms of communication capabilities, without all of the complexity of the NGN architecture.