June 30, 2004
Moving into 3G, operators are faced with the challenge of launching a growing number of new services and then ensuring that they rapidly become profitable.
Adding new services and managing the network infrastructure must be easy. That is why the layered architecture concept will become increasingly important for operators. It offers opportunities to reduce capital and operating expenditure by offering a smooth step-by-step migration to IP, the network technology Ericsson believes will be the foundation for all telecom services, wireline as well as mobile.
This new architecture is future-proof, meaning it will support any kind of service based on IP, where currently all new development is taking place. It also protects previous investments, as operators can choose to migrate present services on a case-by-case basis, right away or later on.
Ericsson is now delivering the layered architecture concept for GSM. The new architecture has already been implemented as standard for WCDMA and it is in operation all over the world in more than ten 3G networks from Ericsson. The new architecture allows operators to share infrastructure for 2G and 3G and supports the migration from 2G to 3G by moving to a common core network.
Layered architecture consists of several horizontal layers that are more or less independent of each other. It differs in several aspects from the traditional vertical architecture in which each radio or wireline network is a completely separate network with its own monitoring, switching, transport and access components. Today, as networks become more tightly integrated with each other, this architecture is becoming very cumbersome to manage.
A layered network is generally composed of:
One of the most significant benefits of the new architecture is that each layer can be scaled, upgraded and optimized separately, meaning that the operator can change the technology used in the transport layer, for example, from TDM to ATM and IP without noticeably affecting the control layer.
Key benefits are that network resources can be used more effectively in terms of simplicity and fewer equipment sites leading to lower total cost of ownership. Also, the need for transmission connections in the network can, in many cases, be reduced by more than 50 percent. The architecture supports local switching, meaning that a call between two people located in the same geographic area does not need to be passed upward through the core network. Multi-vendor implementations are also supported and it will be easier to introduce new functions and services into the networks.