BAP

Backhaul Adaptation Protocol

Protocol →
Introduced in Rel-16 Also in: Radio Access Network

BAP is the layer 2 protocol for 3GPP IAB networks that dynamically routes data and control packets between IAB nodes and the donor node over wireless backhaul links.

Category
Protocol
Introduced
Rel-16
Where
Security
Also touches
1 segments
Specifications
5 specs
BAP Description Purpose Detected Changes Specifications

Description

The Backhaul Adaptation Protocol (BAP) is a key protocol layer within the IAB architecture, operating above the Radio Link Control (RLC) layer and below the Service Data Adaptation Protocol (SDAP) and Packet Data Convergence Protocol (PDCP) layers for user plane traffic. Its primary function is to provide routing and bearer mapping across the multi-hop wireless backhaul formed by IAB nodes. Each IAB node and the IAB donor contain a BAP entity. The BAP layer adds a BAP header to packets, which contains a BAP Routing ID. This ID is used to route the packet along the correct path through the backhaul topology towards its final destination, which could be another IAB node, the donor, or a UE attached to an IAB node.

BAP operates with two types of bearers: BAP bearers and radio bearers. A BAP bearer represents an end-to-end logical connection between the donor node's central unit (CU) and an IAB node or a UE served by an IAB node. It is identified by a BAP address and a BAP path ID. The BAP layer maps these BAP bearers onto radio bearers (e.g., RLC channels) on each individual wireless hop. This mapping is configurable and allows for traffic differentiation and QoS handling across the backhaul. The routing tables within each BAP entity, which map BAP Routing IDs to the next-hop link, are configured by the IAB donor's CU via F1-Application Protocol (F1-AP) signaling over the control plane.

For downstream traffic (from network to UE), the donor CU determines the BAP Routing ID. As the packet traverses each IAB node, the local BAP entity examines the BAP Routing ID, consults its routing table, and forwards the packet to the correct next-hop child node via the appropriate radio bearer. For upstream traffic, the IAB node where the UE is attached adds the BAP Routing ID. This ID is typically configured by the donor and directs the packet along the upstream path towards the donor. BAP also supports topology adaptation. When a link fails or a node is added, the donor CU can reconfigure the BAP routing tables in the affected nodes to establish new paths, enabling robust and self-healing backhaul networks.

Purpose & Motivation

BAP was created to address the challenge of building scalable and flexible wireless backhaul for 5G networks, specifically for the Integrated Access and Backhaul (IAB) feature. Traditional wired or point-to-point microwave backhaul is expensive and inflexible for dense, small-cell deployments. IAB allows 5G base stations (gNBs) to use part of their radio resources for backhauling traffic from other, more remote nodes, creating a wireless mesh. However, this required a new protocol to manage multi-hop routing within the gNB architecture without involving the core network.

The purpose of BAP is to provide a layer 2 routing mechanism that is tightly integrated with the 3GPP NG-RAN architecture. Prior to BAP, multi-hop networking would require IP routing at layer 3, which adds complexity, overhead, and is less optimal for the latency-sensitive and tightly synchronized RAN environment. BAP solves this by operating below the PDCP layer, allowing the IAB network to appear as a single, logical gNB to the 5G core. It enables efficient hop-by-hop forwarding, supports QoS differentiation across the backhaul by mapping to different RLC channels, and facilitates fast topology management under the control of the donor CU. This allows operators to rapidly deploy coverage, especially in areas where fiber is not available, by daisy-chaining nodes wirelessly.

Release Timeline

Detected Changes Across Releases

from 3GPP Change Requests

Specific changes extracted from the „Change history“ tables of 3GPP specifications (2 CRs across 2 releases). Complements the general historical overview above with the evidence-based evolution of this function.

Rel-16 1 change

In Release 16, specific corrections and refinements were made to the Backhaul Adaptation Protocol (BAP) to improve its configuration. These changes included addressing issues related to BAP addressing and the setup of the default BAP configuration. This ensured the proper establishment of BH RLC channels and the BAP layer via RRC signaling from the IAB-donor for routing between the IAB-node and the IAB-donor.

  • Corrections on BAP address and default BAP configuration TS 38.331CR2427
Rel-17 1 change

In Release 17, a specific correction was made to the Backhaul Adaptation Protocol (BAP) concerning its entity release procedures during Multi-Radio Dual Connectivity (MR-DC) release, as detailed in the specification TS 38.331. This update ensures proper handling when an IAB-node's connection to the IAB-donor is reconfigured or terminated. The change refines the coordination between BAP layer management and the higher-layer RRC signaling used for backhaul RLC channel establishment.

  • Corrections on BAP entity release in MR DC release procedures in TS 38.331 TS 38.331CR3061

Explore further

Broader topics and technologies where BAP plays a role.

Defining Specifications

3GPP specifications that define or reference BAP, with the latest known release. Sourced from the 3GPP document catalog — see methodology.

SpecificationTitleRelease
TS 33.501 vk00 5G Security Architecture and Procedures Rel-20
TR 33.824 vh00 Security Study for NR Integrated Access & Backhaul Rel-17
TS 38.306 vj00 NR UE Radio Access Capability Parameters Rel-19
TS 38.331 vj00 NR Radio Resource Control (RRC) Protocol Specification Rel-19
TS 38.401 vj10 NG-RAN Architecture Specification Rel-19
Patrick Zandl

About the author: Patrick Zandl (b. 1974)

Telecommunications specialist, technology journalist (founder of the Mobil server), and developer who has been running since 2025 — the largest Czech-language resource on AI-assisted programming. Formerly Chief Wizard Architect at Prusa3D and head of development for Turris at CZ.NIC; currently a consultant and instructor on AI implementation in companies.