Description
Dual Connectivity (DC) is a fundamental radio resource management feature in 3GPP standards that allows a single User Equipment (UE) to maintain concurrent connections with two distinct network nodes, referred to as a Master Node (MN) and a Secondary Node (SN). These nodes can belong to the same or different radio access technologies (RATs), such as LTE (E-UTRA) and NR (New Radio). The UE is configured with two separate protocol stacks, one for each node, enabling the simultaneous transmission and reception of data over both radio links. The MN provides the control plane connection to the core network (e.g., via S1-MME or NG-C) and manages the overall UE context, including the establishment, modification, and release of the secondary connection. The SN provides additional radio resources for the user plane, boosting capacity. Data can be split at various protocol layers (e.g., PDCP, RLC) depending on the DC architecture variant (e.g., MCG bearer, SCG bearer, split bearer).
The architecture is defined by the roles of the involved nodes. In LTE DC (introduced as LTE-NR Dual Connectivity or EN-DC), the eNB typically acts as the MN (MeNB), and a gNB acts as the SN (SgNB). In 5G NR DC (NR-NR DC or NE-DC), the roles are defined as MN (e.g., a gNB) and SN (another gNB). The nodes are interconnected via standardized interfaces: the X2 interface for LTE-based nodes and the Xn interface for NR-based nodes. These interfaces carry control plane signaling (e.g., SN Addition/Modification/Release procedures) and user plane data for bearers terminated at the SN. The UE maintains two Cell Groups: the Master Cell Group (MCG) associated with the MN and the Secondary Cell Group (SCG) associated with the SN. Each group comprises a Primary Cell (PCell or PSCell) and optionally one or more Secondary Cells (SCells).
Operationally, DC involves complex coordination. The MN makes the decision to add an SN based on measurement reports from the UE and its own load conditions. It initiates the SN addition procedure via the X2/Xn interface, transferring necessary UE context. The SN then performs its own admission control and, if successful, configures resources for the UE. The MN provides the final configuration to the UE via RRC signaling, which may include a secondary RRC configuration from the SN (in the case of NR DC). For user plane, data can be routed in different ways. In a split bearer configuration, the PDCP layer at the MN handles packet duplication, sequencing, and can route packets to either its own RLC layer (for the MCG leg) or to the SN's RLC layer (for the SCG leg) via the X2/Xn-U interface. This requires tight synchronization and flow control between the nodes to minimize packet reordering delays at the receiver.
DC's role in the network is multi-faceted. Primarily, it is a capacity-boosting tool, aggregating spectrum and radio resources from two transmission points to achieve peak data rates beyond what a single node can provide. It is also a critical mobility enhancement; by keeping an anchor connection (the MCG) stable, it allows for smoother handovers of the SCG, reducing the risk of radio link failure during inter-cell mobility. Furthermore, it enables efficient load balancing between different network layers (e.g., macro and small cells) or different frequency bands. In 5G, DC is the foundation for more advanced multi-connectivity schemes and is essential for leveraging non-standalone (NSA) architectures where the LTE anchor provides robust coverage and control, while the NR link delivers high throughput.
Purpose & Motivation
Dual Connectivity was created to address the growing demand for higher user data rates and more robust mobile experiences, which could not be met by a single connection to one network cell. Prior to DC, Carrier Aggregation (CA) allowed a UE to aggregate component carriers from a single base station, but this was limited by the geographical coverage and capacity of that one node. DC overcomes this limitation by allowing aggregation across geographically separated nodes, effectively pooling the resources of a macro cell and a small cell. This was particularly important for heterogeneous network (HetNet) deployments, where small cells are deployed to boost capacity in hotspots but require a stable macro cell layer for control and coverage.
Historically, the concept evolved from earlier multi-point coordination techniques. Its formal introduction in 3GPP Release 12 (for LTE-LTE DC) was driven by the need for improved per-user throughput and mobility performance in dense networks. A key problem it solved was the 'ping-pong' effect in small cell deployments, where frequent handovers could degrade performance. By anchoring the control plane at the macro cell (MN) and adding a small cell as an SN for data, DC provided a stable connection while dynamically adding and removing capacity. This also improved network energy efficiency by allowing the SN to be activated only when needed for high data traffic.
The motivation intensified with the advent of 5G. The initial 5G deployments used the Non-Standalone (NSA) architecture, which relied fundamentally on LTE-NR Dual Connectivity (EN-DC) to provide a 5G data pipe (via the NR gNB as SN) while maintaining the LTE eNB as the control anchor. This allowed for rapid 5G service rollout using existing LTE core networks. Furthermore, as 5G networks evolved to standalone (SA) mode, NR-NR DC (and later multi-RAT DC) became essential for aggregating diverse NR frequency ranges (e.g., FR1 and FR2/mmWave) to combine coverage and capacity, ensuring consistent high performance even when one link (like mmWave) is susceptible to blockages.
Key Features
- Simultaneous connectivity to a Master Node and a Secondary Node
- User plane aggregation for increased peak data rates
- Control plane anchoring at the Master Node for robustness
- Support for multiple RAT combinations (e.g., LTE-LTE, LTE-NR, NR-NR)
- Flexible bearer types (MCG, SCG, and Split Bearer) for data routing optimization
- Enhanced mobility and load balancing across network layers
Evolution Across Releases
Introduced the foundational Dual Connectivity framework for LTE, specifying the architecture where a UE connects to a Master eNB (MeNB) and a Secondary eNB (SeNB). Defined three bearer types: MCG bearer, SCG bearer, and split bearer. Established control procedures over the X2 interface and specified the user plane protocol stack with data split at the PDCP layer for split bearers.
Introduced 5G NR and defined key DC architectures for 5G deployment: EN-DC (LTE eNB as MN, NR gNB as SN), NE-DC (NR gNB as MN, LTE eNB as SN), and NR-NR DC. Standardized the use of the Xn interface for inter-gNB communication. Enhanced the framework to support the new 5G core network (5GC) and the split of the RRC protocol, allowing the SN to provide its own RRC configuration in NR-NR DC.
Enhanced DC with features like conditional SN addition/change, improving mobility robustness. Introduced support for dual connectivity with multi-RAT Dual Connectivity (MR-DC) enhancements, including tighter integration with Carrier Aggregation. Added improvements for uplink transmission, power control, and measurements to support more complex deployments and frequency ranges.
Further refined MR-DC operations, focusing on performance enhancements for integrated access and backhaul (IAB) nodes operating in DC. Introduced enhancements for multi-TRP (Transmission Reception Point) operation, leveraging DC principles for improved reliability and reduced latency. Extended support for non-terrestrial networks (NTN) in a DC context.
Continued evolution towards advanced multi-connectivity, exploring tighter inter-node coordination for joint transmission/reception. Focused on enhancements for extreme high-frequency bands (e.g., upper mmWave) and AI/ML-assisted management of DC connections for optimized resource utilization and energy efficiency in the network.
Defining Specifications
| Specification | Title |
|---|---|
| TS 21.905 | 3GPP TS 21.905 |
| TS 23.110 | 3GPP TS 23.110 |
| TS 23.228 | 3GPP TS 23.228 |
| TS 23.334 | 3GPP TS 23.334 |
| TS 23.392 | 3GPP TS 23.392 |
| TS 23.700 | 3GPP TS 23.700 |
| TS 23.725 | 3GPP TS 23.725 |
| TS 24.186 | 3GPP TS 24.186 |
| TS 24.392 | 3GPP TS 24.392 |
| TS 25.113 | 3GPP TS 25.113 |
| TS 25.301 | 3GPP TS 25.301 |
| TS 25.302 | 3GPP TS 25.302 |
| TS 25.321 | 3GPP TS 25.321 |
| TS 25.322 | 3GPP TS 25.322 |
| TS 25.707 | 3GPP TS 25.707 |
| TS 26.264 | 3GPP TS 26.264 |
| TS 26.567 | 3GPP TS 26.567 |
| TS 26.927 | 3GPP TS 26.927 |
| TS 26.982 | 3GPP TS 26.982 |
| TS 26.998 | 3GPP TS 26.998 |
| TS 28.802 | 3GPP TS 28.802 |
| TS 28.851 | 3GPP TS 28.851 |
| TS 29.175 | 3GPP TS 29.175 |
| TS 29.330 | 3GPP TS 29.330 |
| TS 29.364 | 3GPP TS 29.364 |
| TS 32.158 | 3GPP TR 32.158 |
| TS 32.300 | 3GPP TR 32.300 |
| TS 32.425 | 3GPP TR 32.425 |
| TS 33.128 | 3GPP TR 33.128 |
| TS 33.790 | 3GPP TR 33.790 |
| TS 33.825 | 3GPP TR 33.825 |
| TS 33.890 | 3GPP TR 33.890 |
| TS 34.124 | 3GPP TR 34.124 |
| TS 36.101 | 3GPP TR 36.101 |
| TS 36.104 | 3GPP TR 36.104 |
| TS 36.113 | 3GPP TR 36.113 |
| TS 36.116 | 3GPP TR 36.116 |
| TS 36.117 | 3GPP TR 36.117 |
| TS 36.124 | 3GPP TR 36.124 |
| TS 36.141 | 3GPP TR 36.141 |
| TS 36.300 | 3GPP TR 36.300 |
| TS 36.302 | 3GPP TR 36.302 |
| TS 36.306 | 3GPP TR 36.306 |
| TS 36.331 | 3GPP TR 36.331 |
| TS 36.410 | 3GPP TR 36.410 |
| TS 36.423 | 3GPP TR 36.423 |
| TS 36.424 | 3GPP TR 36.424 |
| TS 36.425 | 3GPP TR 36.425 |
| TS 36.770 | 3GPP TR 36.770 |
| TS 36.875 | 3GPP TR 36.875 |
| TS 36.876 | 3GPP TR 36.876 |
| TS 36.894 | 3GPP TR 36.894 |
| TS 37.113 | 3GPP TR 37.113 |
| TS 37.141 | 3GPP TR 37.141 |
| TS 37.340 | 3GPP TR 37.340 |
| TS 37.461 | 3GPP TR 37.461 |
| TS 37.716 | 3GPP TR 37.716 |
| TS 37.717 | 3GPP TR 37.717 |
| TS 37.718 | 3GPP TR 37.718 |
| TS 37.719 | 3GPP TR 37.719 |
| TS 37.802 | 3GPP TR 37.802 |
| TS 37.825 | 3GPP TR 37.825 |
| TS 37.863 | 3GPP TR 37.863 |
| TS 37.872 | 3GPP TR 37.872 |
| TS 37.878 | 3GPP TR 37.878 |
| TS 37.898 | 3GPP TR 37.898 |
| TS 37.900 | 3GPP TR 37.900 |
| TS 37.910 | 3GPP TR 37.910 |
| TS 38.101 | 3GPP TR 38.101 |
| TS 38.113 | 3GPP TR 38.113 |
| TS 38.114 | 3GPP TR 38.114 |
| TS 38.124 | 3GPP TR 38.124 |
| TS 38.133 | 3GPP TR 38.133 |
| TS 38.171 | 3GPP TR 38.171 |
| TS 38.175 | 3GPP TR 38.175 |
| TS 38.202 | 3GPP TR 38.202 |
| TS 38.213 | 3GPP TR 38.213 |
| TS 38.331 | 3GPP TR 38.331 |
| TS 38.413 | 3GPP TR 38.413 |
| TS 38.521 | 3GPP TR 38.521 |
| TS 38.522 | 3GPP TR 38.522 |
| TS 38.716 | 3GPP TR 38.716 |
| TS 38.717 | 3GPP TR 38.717 |
| TS 38.718 | 3GPP TR 38.718 |
| TS 38.719 | 3GPP TR 38.719 |
| TS 38.746 | 3GPP TR 38.746 |
| TS 38.750 | 3GPP TR 38.750 |
| TS 38.755 | 3GPP TR 38.755 |
| TS 38.792 | 3GPP TR 38.792 |
| TS 38.793 | 3GPP TR 38.793 |
| TS 38.802 | 3GPP TR 38.802 |
| TS 38.804 | 3GPP TR 38.804 |
| TS 38.820 | 3GPP TR 38.820 |
| TS 38.825 | 3GPP TR 38.825 |
| TS 38.839 | 3GPP TR 38.839 |
| TS 38.841 | 3GPP TR 38.841 |
| TS 38.842 | 3GPP TR 38.842 |
| TS 38.846 | 3GPP TR 38.846 |
| TS 38.873 | 3GPP TR 38.873 |
| TS 38.880 | 3GPP TR 38.880 |
| TS 38.881 | 3GPP TR 38.881 |
| TS 38.889 | 3GPP TR 38.889 |
| TS 38.894 | 3GPP TR 38.894 |
| TS 38.899 | 3GPP TR 38.899 |
| TS 38.912 | 3GPP TR 38.912 |
| TS 43.051 | 3GPP TR 43.051 |
| TS 44.060 | 3GPP TR 44.060 |