Architecture Analysis of full IP Next Generation Network

1. Introduction

People gradually realize that the next generation of fixed and mobile phone network architecture will be based on IP addresses. The primary motivation for proposing a full IP network is that it is currently the best architecture choice to support the vast majority of innovative and profitable businesses. The second reason is that the basic IP address is relatively simple in technology. The IP address itself has a unified extraction function, which can hide and share the complexity of the protocol stack, it also makes it easy to develop IP network applications. This will lead to the rapid development and deployment of a large number of Internet applications. Some of these applications (such as E-mail, short messages, and content delivery) can directly compete with the current or potential market of the PSTN network. The second reason is that with the emergence of new Internet applications, the telecom network must be effectively integrated with the Internet, or allow users to access Internet applications through the network, it either allows interoperation with Internet applications in some aspects, and IP-based next-generation network architecture applications can make this integration easier.

2. API architecture of the Next Generation Network

In a sense, the next generation network must be a programmable IP-based network, which requires XG to provide powerful, convenient, and clearly defined APIs for application development. For some service providers, the API concept is not just a basic concept. Currently, public APIs are APIs available to all third parties, including standard and open API subsets. On the other hand, private APIs are APIs controlled by a company that are only available within the company or within the partner.

From a historical perspective, the telecom network is composed of a complex, intelligent core network and a simple terminal. The core network is highly integrated to meet strict performance and reliability requirements. Terminal devices can be provided by the market in large quantities and provide fixed and easy-to-use user interfaces. This is in stark contrast to the Internet. The Internet mainly considers connecting very complex terminal devices through a relatively simple core router, leading to a decline in the overall performance and reliability of the system. The compromise between the smart core network and the simple core network has attracted wide attention and become a key decision in the architecture.

IN the past 10 years, the intelligence of core network switching and routing functions has been decreasing. IN 1970s, the introduction of common channel signaling (CCS), intelligent network (IN) the emergence of the architecture has intensified this trend. In the intelligent network architecture, the business logic of the Telecom switch is removed and moved to the distributed database.

IN fact, the SS7 and IN architectures mark the emergence of the control layer. The control layer is becoming more mature and more distributed. The design of the Next Generation Network (NGN) in 1990s may be called the Group voice (VoP). It uses the group transmission network (usually ATM rather than the IP network ), this network is controlled by a high-level call proxy (or Softswitch) responsible for all call processing. The call proxy also provides business logic and interfaces for external application servers. The updated architecture requires that you directly use the Internet protocol (such as the SIP protocol and Mobile IP protocol) to configure features on distributed servers (such as local proxies and registers.

On the other hand, the use of overlapping networks in IP networks is growing steadily. It can be used as an optional solution for configuring and using core router functions. This is mainly determined by the following factors: the standardization process of new protocols lags behind; the complexity of core elements (routers) is constantly improved, and the openness is lacking; EDGE network computing is very economical. Interestingly, these are all shortcomings of traditional telecom networks. Similarly, the intelligence of mobile terminals and Wireless Access Networks (RAN) is constantly improved. Compared with the centralized design of base stations and base station controllers, the control functions are gradually distributed. The next generation network architecture conforms to these historical development trends.

3. layered structure of the Next Generation Network

The hierarchy of the next-generation network with full IP addresses includes the stack layer, control layer, core layer, and access layer. The next-generation network architecture includes a core IP network, which is extremely intelligent. Therefore, most of the core network functions (such as Routing) are implemented by existing and emerging IP technologies. On the core network, an advanced control layer cannot provide routing and call path creation functions. The advanced control layer transfers these functions to the core network. The advanced control layer focuses on the functions that can be used for applications and network elements, such as access to AAA decision points, Mobility Management Agents, role and rules of policy management. The sparse coupling between the control layer and the core network means that the advanced control layer is usually not involved in the fast path creation process of group forwarding and processing. The core network is a collection of access networks used to meet different market opportunities and needs. 4g ran is the product of the evolution of the current Wireless Access Network (RAN) to a higher data rate. It supports interoperability services, multimedia services, and distributed control elements that are interconnected through IP networks. As real-time restrictions are critical to this layer, strict coordination and coupling are required between the core network and the access network. The core network also provides support services and connections for specialized networks, such as enterprise networks, multi-hop/Ad Hoc networks, and peer-to-peer networks operated by next-generation network operators. These specialized networks may require local control, especially for key features (such as AAA. The provision of anonymous services in some regions, while ensuring the quality of service (QoS) and reliability, is a key issue that needs to be studied at present.

The next-generation network architecture has a stack layer that provides high-level functions and services for applications, such as application layer multicast (ALM), location services, and content delivery services. The stack layer can be divided into two levels: low-level features (such as ALM) and high-level features (such as location services) that are close to the core network ).

4. function structure of the Next Generation Network

If four abstract layers in the horizontal direction and their functions are subdivided, the functions can be divided into multiple vertical sets. Each set is called a "surface", including the key capabilities that span all layers or across several layers. These aspects are security, QoS/resource control, and other similar functions (such as transmission, mobility, networking, and business control ). Independent Parallel surfaces include operation, management, maintenance, and user equipment surfaces. Each plane and each layer are independent from each other to form an object-oriented network architecture, which is easy to maintain and upgrade.

The low layer (L1, L2, and L2.5) is an access network layer that provides physical and Media Access Control (MAC) connections, necessary access control, wide area mobility, and QoS-guaranteed switching capabilities. The highest layer is the IP-based access network, which provides IP connectivity, necessary access control, integrated QoS Management, address allocation, and the ability to switch between subnets using the fast Mobile IP protocol. The two layers are flexible and mixed in different combinations. The combination mainly depends on the access network technology and the specific topology requirements of a certain part of the network.

The core network layer consists of pure IP addresses that distinguish between core networks. These networks generally provide original bandwidth for connecting different parts of the network. It also includes gateways used to connect external networks (such as the Internet) and uses necessary preventive measures to prevent DoS attacks from external networks ).

Network services that assist other layers in their tasks and missions are called support services. The supported services can be divided into two levels. Level 1 support services are mostly related to the network transmission function; Level 2 Support Services provide necessary functions for terminal services to work normally. Level 1 businesses include network-level AAA businesses, roaming businesses, macro mobility management businesses, and QoS execution functions, qoS execution is mainly used to configure different parts of the network to provide the service quality that conforms to the network policy and user configuration file.

Level 2 support businesses provides a rich set of services that facilitate the development of terminal businesses, including allowing applications to discover and interact with businesses to provide more advanced combined businesses for business registration. This layer provides the application layer AAA service for Terminal Services, and facilitates application and content distribution and other optimization services for overlapping networks, Certificate Services, and a series of gateways, these gateways can provide business-level integration with traditional network services (such as voice, video, and PSTN networking services in 3G networks.

It should be noted that IP addresses are used as basic transmission mechanisms in all layers (including wireless access networks and service layers). Generally, it is assumed that IP-based next-generation networks and traditional networks (or even Internet connections) are connected) the correct gateway and firewall are used. At the same time, function interconnection or physical interconnection may be cross-layer (such as from end user services to core networks). This cross-layer connection facilitates cross-layer improvement through cross-layer APIs.

5. Conclusion

Currently, bandwidth-consuming services such as IPTV will require a new IP infrastructure. Carriers also want to reduce operating costs by turning all services to IP networks. The full IP architecture is the best choice for developing innovative and profitable businesses. The Next Generation Network allows operators to call service platforms and IP Multimedia subsystems on the basis of the new network, integrate various operational services, and launch various new services more quickly. In general, global network operators will slowly start to promote the next generation network of all IP addresses. For a fixed-line network, switching to an IP-based infrastructure is the natural evolution of broadband services, including IP phones, which will replace the current telephone network. The mobile network will switch to the IP address structure at a slow speed at the same time, but the standardization work will speed up.

1. Next Generation Network (NGN): Research Progress of Softswitch Technology
2. IPv6 and next-generation networks
3. Topic: NGN