Mobile Core Network Architecture Analysis

Overview of Mobile Core Network

The mobile core network (Mobile Core Network) is the central nervous system of the mobile communication system, responsible for handling core functions such as user authentication, session management, mobility management, and policy control. It connects the Radio Access Network (RAN) with external networks (such as the Internet and enterprise private networks), providing end-to-end communication services to mobile users. The architecture design of the core network directly affects the network's performance, reliability, scalability, and service capabilities, serving as the foundational support platform for the entire mobile communication network.

With the evolution of mobile communication technology, the core network architecture has also been continuously developing. From the circuit-switched network of the 2G era, to the packet-switched network of 3G, the flat full-IP network of 4G, and the service-oriented architecture of 5G, each generation of technology has brought new requirements for the core network. Modern mobile core networks must not only support traditional voice and data services but also meet the diverse needs of emerging applications such as the Internet of Things, industrial Internet, and vehicle networks, achieving a transition from single-service platforms to multi-service integrated platforms.

The development of core network technology reflects the trend of communication networks evolving toward intelligence, cloudification, and service orientation. Traditional dedicated hardware devices are gradually being replaced by general-purpose servers and cloud platforms. Network functions are shifting from rigid architectures to flexible orchestration, and service deployment is moving from periodic planning to dynamic on-demand provision. This transformation not only enhances network flexibility and resource utilization but also provides architectural support for new technologies in the 5G era, such as network slicing and edge computing.

Evolution of Core Network Technology

4G Core Network EPC Architecture

The 4G LTE network adopts the Evolved Packet Core (EPC) architecture, which is a flat, all-IP based network architecture. EPC mainly consists of four core network elements: MME (Mobility Management Entity), SGW (Serving Gateway), PGW (PDN Gateway), and HSS (Home Subscriber Server). This architecture abandons the traditional circuit-switched domain, uniformly uses packet-switched technology, simplifies network architecture, improves data transmission efficiency, and reduces end-to-end latency.

The MME is the control plane core of the EPC, responsible for handling user mobility management, session management, security management, and other functions. It manages user location information, handles signaling processes such as attach, detach, and tracking area updates, and is also responsible for user authentication and key management. The SGW is the anchor point of the user plane, responsible for routing and forwarding user data, maintaining continuity of the data path during base station handovers. The PGW is the interface between the EPC and external networks, responsible for IP address allocation, policy enforcement, charging, and supports aggregation of multiple access technologies. The HSS stores users' subscription data and authentication information, providing user data query services to other network elements.

5G Core Network (5GC) Architecture

The 5G Core Network (5GC) adopts a brand-new Service-Based Architecture (SBA), which is a cloud-native distributed architecture. Unlike the element-based architecture of the 4G EPC, 5GC decomposes network functions into finer-grained Network Functions (NFs), which interact with each other through standardized service interfaces, achieving better modularity, scalability, and flexibility.

The main network functions of 5GC include AMF (Access and Mobility Management Function), SMF (Session Management Function), UPF (User Plane Function), UDM (Unified Data Management), AUSF (Authentication Server Function), PCF (Policy Control Function), etc. AMF is responsible for registration management, connection management, mobility management, and access authentication; SMF is responsible for session management, IP address allocation, UPF selection, and so on; UPF is the user plane anchor, responsible for packet routing and forwarding, QoS processing, usage reporting, and more; UDM centrally manages user data, including subscription data, policy data, etc.; AUSF handles user authentication; PCF is responsible for policy decision-making and control.

Network Slicing Technology

Network slicing is an important feature of the 5G core network, allowing the creation of multiple logically isolated network instances on the same physical network infrastructure, each of which can be customized according to specific service requirements. Different network slices can have different combinations of network functions, QoS parameters, security policies, and service level agreements, thereby meeting the differentiated needs of various vertical industries and application scenarios. This technology realizes the vision of 'one network, multiple services.'

The implementation of network slicing involves the coordination of the core network, transport network, and access network. On the core network side, customized control plane and user plane functions are provided for different slices through flexible combination and instantiation of network functions. On the transport network side, technologies such as FlexE and segment routing are used to achieve resource isolation and guarantee. On the access network side, wireless resource isolation and allocation are achieved through RAN slicing technology. The network slice manager is responsible for the lifecycle management of slices, including creation, configuration, monitoring, modification, and deletion of slices.

Mobility Management

Mobility management is one of the core functions of the core network, ensuring users can maintain continuity of communication while moving. In 4G networks, the MME is responsible for mobility management, including tracking area (TA) management, paging, handover, and other functions. When a user moves within the network, the MME tracks the user's location information and initiates a paging process to locate the user when necessary. During handover between base stations, the MME coordinates related network elements to complete path switching, ensuring continuity of data transmission.

Mobility management in 5G networks is more complex and flexible. The AMF is responsible for mobility management, supporting finer-grained mobility control. 5G introduces the concept of a registration area (RA), allowing users to move freely within the area without frequently updating their location. The SMF is responsible for PDU session mobility management, supporting session establishment, modification, and release. The UPF serves as the user plane anchor, maintaining data path continuity during mobility. 5G also supports multi-access and multi-connectivity technologies, allowing users to connect to multiple access points simultaneously for a better mobility experience.

Session Management and QoS Control

Session management is responsible for establishing, maintaining, and releasing user data connections. In 4G networks, MME and SGW work together to perform bearer management, establishing end-to-end data transmission channels for users. Each bearer has specific QoS parameters, including QCI (QoS Class Identifier), ARP (Allocation and Retention Priority), GBR (Guaranteed Bit Rate), MBR (Maximum Bit Rate), etc. PGW allocates appropriate QoS parameters for different service flows according to user subscription information and policy rules.

Session management in 5G networks is more flexible and detailed. SMF is responsible for PDU session management and supports multiple PDU types such as IPv4, IPv6, and Ethernet. 5G introduces new QoS parameters such as 5QI (5G QoS Identifier), GFBR (Guaranteed Flow Bit Rate), MFBR (Maximum Flow Bit Rate), which better support the QoS requirements of different services. Network slicing technology allows different slices to have independent QoS policies, enabling differentiated service assurance. PCF is responsible for policy decisions, dynamically adjusting QoS parameters based on user subscriptions, network conditions, service requirements, and other factors.

Security Architecture and Authentication

Security is a major consideration for the mobile core network, involving multiple aspects such as user authentication, data encryption, and network access control. The security architecture of 4G networks is based on the AKA (Authentication and Key Agreement) protocol, ensuring the trustworthiness of user identity through mutual authentication between the HSS and the user's SIM card. Network domain security protects the communication between network elements using technologies like IPSec. User data is encrypted during transmission over both the air interface and within the core network.

5G networks, building on 4G security mechanisms, further enhance security capabilities. The AUSF is specifically responsible for user authentication and supports authentication methods such as 5G-AKA and EAP-AKA'. SEAF (Security Anchor Function) handles the security functions at the access layer. UDM stores authentication-related security information. 5G also introduces new security features such as network slicing isolation, edge computing security, and user plane integrity protection, constructing a more comprehensive security protection system. Dynamic configuration of security policies and real-time monitoring of security events have become key characteristics of 5G network security management.

Network Function Virtualization

Network Function Virtualization (NFV) is an important feature of modern core network architecture. Traditional core network elements mostly rely on dedicated hardware, which is difficult to upgrade and costly. NFV technology decouples network functions from dedicated hardware and runs them on general-purpose servers and cloud platforms, achieving the software-based and virtualized implementation of network functions. This architecture enhances network flexibility, scalability, and cost-effectiveness.

In the NFV architecture, a VNF (Virtual Network Function) is a software implementation of core network functions, running on virtual machines or containers. The NFVI (NFV Infrastructure) provides basic resources such as computing, storage, and networking. MANO (Management and Orchestration) is responsible for the lifecycle management of VNFs and resource orchestration. 5G networks further introduce cloud-native technologies, adopting a microservices architecture and containerized deployment to achieve better resource utilization efficiency and service agility. The dynamic creation and management of network slices largely rely on the support of NFV and cloud-native technologies.

IoT Core Network Optimization

IoT applications pose new challenges and requirements for the core network. The massive access of IoT devices requires the core network to have stronger concurrency processing capabilities. The diversity of IoT devices demands that the core network supports different access methods and communication modes. Requirements for low power and low cost drive the simplification and optimization of core network functions. IoT-specific technologies such as NB-IoT and eMTC require dedicated optimization support on the core network side.

For IoT applications, the core network has undergone multiple optimizations. Signaling optimization reduces device attachment and detachment overhead, supporting long-cycle sleep modes. Data transmission optimization supports efficient transmission of small data packets, reducing protocol overhead. Mobility management optimization simplifies the location update process for stationary or low-mobility devices. Security optimization reduces the complexity of authentication and encryption while ensuring security. These optimization measures enable the core network to better serve the special requirements of IoT applications.

Edge Computing and Core Network

The rise of edge computing has changed the architectural model of the traditional core network. In traditional architectures, all data needs to be backhauled to a centralized core network for processing, which leads to higher latency and bandwidth consumption. Edge computing achieves local data processing by deploying computing and storage resources at the network edge, significantly reducing latency and alleviating the backbone network load. This change has a profound impact on core network architecture.

The 5G core network supports edge computing through flexible deployment of UPF. UPF can be deployed at the network edge, forming local data offloading points that enable local service processing. Network slicing technology provides customized network services for edge computing, allowing different edge applications to use dedicated network slices. A service-based architecture allows network functions to be flexibly deployed in different locations, including central cloud, edge cloud, and local data centers. This distributed architectural model provides technical support for low-latency applications, localized services, and privacy protection.

Future Development Trends

The mobile core network is evolving toward intelligence, service-orientation, and cloudification. The introduction of artificial intelligence technology will enable the core network to have self-learning and self-optimization capabilities, achieving intelligent scheduling of network resources and automatic fault handling through machine learning algorithms. Intent-Based Networking (IBN) will simplify network configuration and management. Operators only need to describe the business intent, and the network system can automatically complete the corresponding configuration and optimization.

In the 6G era, the core network will further evolve to support more diverse business requirements and application scenarios. The integrated space-air-ground-sea network will require the core network to support the convergence of multiple access technologies. Digital twin technology will provide new methods for network planning, optimization, and management. Emerging technologies such as blockchain and quantum communication will also impact core network architecture. The development of these technologies will drive the core network from a communication infrastructure to an intelligent service platform, providing strong network support for the future digital society.