Ericsson ahead in layered architecture

News Archive

Today, an increasing number of operators are choosing to lower capital expenditure and daily costs through layered architecture, where networks are split into horizontal rather than vertical layers. Ericsson, having performed more implementations of this kind than any other vendor, is leading the field in the introduction of layered architecture for both fixed and mobile networks.

March 1, 2005

In the face of increased competition, both fixed and mobile operators are looking to increase the efficiency of their networks. Mobile operators need to bring down capital and operating expenditure to cope with the increased traffic resulting from falling tariffs. And fixed-line operators want to reduce expenditure so that they can maintain high margins on voice traffic; still their chief source of revenue. Both are therefore looking to layered architecture for a solution, as the network efficiency from horizontal splitting can reduce transmission costs by as much as 60 percent.

The most significant feature of a horizontal network is that the mobile switching center (MSC) is divided into two parts: a control unit (MSC Server) and a transport unit (the Media Gateway). This is according to standards specified by bodies such as the Internet Engineering Task Force and the European Telecommunications Standards Institute. It is also included in the 3G Partnership Project's (3GPP) release 4, with which Ericsson's Layered Architecture core network solution is fully compliant.

In practice, the division of the mobile switching centre into two parts means that local switching becomes possible, so a given payload can travel through a network using the shortest path and most efficient coding. For example, a call between two people located in the same geographic area does not need to be passed upward through the entire network. The efficiency of this type of transmission, which renders less traffic through the core of a network, leads to savings in terms of rent, power and maintenance costs.

An additional benefit of layered architecture is that a network's existing platforms and nodes are reused, which allows for maximum protection of operators' investments. Furthermore, layered architecture can be introduced in stages. This means that each network layer can be enhanced, upgraded and optimized separately, so that an operator can, for instance, change the technology used in the transport layer without noticeably affecting the control layer.

As well as facilitating cost savings and protecting existing investments, the separation of the control and connectivity functions is also a vital step towards an all-IP network. Through implementing this architectural change it is possible for operators to achieve convergence between the circuit-switched and packet-switched components of the network; a necessity for smooth evolution to all-IP.

Since 2001, Ericsson has delivered more than 25 layered architecture networks on the fixed side and 17 on the mobile side. In fact, in March 2003 Ericsson performed the first ever implementation of layered architecture for a mobile network. With this experience Ericsson is ahead of the competition and is now building its third generation of media gateways. Operators benefit from a significant body of knowledge, while securing a flexible transition from existing infrastructure towards a cost-effective, 3GPP-compliant network design.