Description
PDN Address Allocation (PAA) is a critical information element defined in 3GPP specifications, particularly in TS 29.274 for the GPRS Tunneling Protocol (GTP), used to convey the IP address assigned to a User Equipment (UE) for a Packet Data Network (PDN) connection. When a UE attaches to the network and requests PDN connectivity—for example, to access the internet or an IMS service—the core network allocates an IP address and communicates it to the UE via the PAA parameter. This parameter is included in GTP-C (GTP Control Plane) messages, such as Create Session Request/Response, exchanged between the Serving Gateway (SGW), Packet Data Network Gateway (PGW) in 4G, or Session Management Function (SMF) in 5G. The PAA contains the actual address values, which can be an IPv4 address, IPv6 prefix, or both (dual-stack), along with additional fields like address type and length.
Technically, PAA works as part of the PDN connectivity establishment procedure. Upon UE request, the PGW/SMF interacts with dynamic host configuration protocol (DHCP) servers or internal address pools to allocate an IP address based on subscription data and network policies. The allocated address is then encapsulated in the PAA field of a GTP message sent to the SGW/UPF, which forwards it to the Mobility Management Entity (MME) or Access and Mobility Management Function (AMF). Finally, it is delivered to the UE via NAS (Non-Access Stratum) signaling, allowing the UE to configure its network interface. The PAA ensures that both the UE and network nodes (e.g., PGW, SGW) have consistent address information for setting up user plane tunnels (GTP-U) and routing data packets correctly.
The role of PAA in the network is fundamental to IP address management in mobile broadband. It enables seamless mobility by ensuring the UE retains its IP address during handovers, supported by GTP context transfer procedures. PAA also supports advanced features like multiple PDN connections per UE, where each connection may have a different PAA for separate services (e.g., internet and IMS). In 5G systems, PAA is adapted within the Packet Forwarding Control Protocol (PFCP) and service-based interfaces, but the concept remains similar. It integrates with policy control (PCRF/PCF) to assign addresses based on QoS profiles or network slicing requirements, ensuring that specific slices receive appropriate address ranges.
Moreover, PAA facilitates network address translation (NAT) and firewall traversal by providing the external IP address assigned to the UE. Security aspects include protection against address spoofing, as the network validates PAA consistency during session management. The parameter is also used in charging records to associate data usage with specific IP addresses for billing purposes. Overall, PAA is a cornerstone of IP connectivity in 3GPP networks, bridging the gap between cellular signaling and Internet Protocol infrastructure.
Purpose & Motivation
PAA was introduced to standardize the method for allocating and communicating IP addresses to UEs in packet-switched cellular networks, solving the problem of dynamic address assignment in mobile environments. Prior to its definition in 3GPP Release 8 with EPS (Evolved Packet System), address allocation mechanisms were less unified, relying on vendor-specific extensions or separate protocols like DHCP over the air, which could lead to interoperability issues and inefficient resource management. The motivation for PAA stemmed from the transition to all-IP networks in LTE, where every data session requires a unique IP address for routing, and the network must manage address scarcity (especially IPv4) while supporting IPv6 migration.
Historically, in 2G/3G GPRS systems, IP address assignment was handled via GGSN and PDP context activation, but the signaling was not as streamlined. PAA addressed limitations by embedding address information directly within GTP-C messages, reducing signaling overhead and enabling faster session setup. This was crucial for supporting always-on connectivity and seamless service continuity, as UEs could quickly re-establish PDN connections with consistent addresses during mobility events. It also solved the challenge of dual-stack operation, allowing simultaneous IPv4 and IPv6 address allocation to facilitate the transition to IPv6 without disrupting existing services.
Furthermore, PAA enables network operators to implement sophisticated address management policies, such as allocating addresses from different pools based on APN (Access Point Name), subscriber tier, or network slice. This supports traffic segregation, enhanced security, and efficient use of address space. By standardizing PAA across network elements, 3GPP ensured that multi-vendor deployments could interoperate reliably, reducing costs and complexity. Its evolution through subsequent releases has included enhancements for 5G, such as integration with service-based architecture, maintaining its role as a key enabler for IP connectivity in modern mobile networks.
Key Features
- Carries allocated IP address (IPv4, IPv6, or dual-stack) in GTP signaling
- Used during PDN connectivity establishment and modification procedures
- Supports dynamic address assignment from network pools or DHCP
- Enables seamless IP address retention during mobility
- Integrates with policy control for address allocation based on subscription
- Facilitates charging and billing through address association in records
Evolution Across Releases
Introduced PAA as a parameter in GTPv2-C for EPS, defining its encoding and usage in Create Session messages to convey UE IP addresses allocated by the PGW. Initial support included IPv4 address allocation, with basic fields for address type and value, enabling standardized IP address management in LTE networks.
Defining Specifications
| Specification | Title |
|---|---|
| TS 29.274 | 3GPP TS 29.274 |