Skip to main content

Integrated access and backhaul: Why it is essential for mmWave deployments

Integrated access and backhaul: Why it is essential for mmWave deployments

5G is about extraordinary network speeds, low latency and reliability for consumers and industries. To deliver on 5G's promise of super-fast data rates and ultra-low latency, physics dictates that we need to tap into higher frequency millimeter waves.  But with high-band mmWave frequencies, radio propagation is much weaker than at the low bands. Signals are easily blocked by buildings, foliage and even human bodies. This means that networks need to be densified in order to maintain coverage and a reliable user experience.

To meet cell densification requirements, initial 5G mmWave deployments require new fiber or wireless backhaul solutions since fiber penetration does not usually match the required ISD (inter site distance). Typically, mmWave deployments require high density with an inter-site distance between base stations of 200-250 meters. This leads to higher costs, with a significant portion coming from the high rate (around 10 Gbps) backhaul connections required for the new access points. And in many regions, the availability of high-speed backhaul, such as optical fiber, is scarce, expensive and not readily available. Not to mention, fiber trenching and site licensing can be time-consuming exercises. To make mmWave deployments both widely accessible and affordable, one possible solution used by Nokia is IAB (integrated access and backhaul).

IAB is an important Rel-16 feature in 5G New Radio (NR) that enables rapid and cost-effective millimeter wave (mmWave) deployments through self-backhauling in the same spectrum. Wireless self-backhauling uses the same wireless channel for coverage and backhaul connectivity to other base stations, which leads to greater performance, more efficient use of spectrum resources and lowers equipment costs, while also reducing the reliance on the availability of wired backhaul at each access node location. In other words, IAB is a multi-hop approach to network deployment and allows deployment of mmWave base stations with or without fiber backhaul transport. It works by having a fraction of the deployed BSs act as donor nodes, using a fiber/wired connection. The remainder without a wired connection are called IAB nodes. Both types of BSs generate an equivalent cellular coverage area and appear identical to user equipment (UE) in its coverage area. IAB allows operators to leverage their existing mmWave spectrum licenses and have the freedom to deploy separate backhaul where needed, without additional interference analysis that can subject to regulatory review (and potential denial).  

IAB is a compliant “cloud RAN” architecture with the split CU (Centralized Unit) and DU (Distributed Unit). Various functions such as RRM (Radio Resource Management), RRC (Radio Resource Control) and PDCP (Packet Data Convergence Protocol) need to be centralized. The Donor DU is a conventional fiber-fed BS connected to the CU using an F1 interface. The IAB node may serve as a first hop or second hop node. Both donor and IAB nodes also directly support UEs multiplexed with the backhaul Ur interface. The Uu interface is directly between a UE and an IAB or donor node.

An IAB network deployment simulation study @ 39 GHz was performed in the suburban Lincoln Park neighborhood in Chicago using a 3-D ray-tracing propagation model. The figure below shows the plot of BS density as a function of UE load. The notation (15, 30) indicates that out of a total of 45 total BSs, there are 15 donor (a.k.a fiber) nodes and 30 IAB nodes. Each line in the plot is for a fixed number of fiber-connected donor nodes (4, 8, 15, 20 or 30). The data rate per UE for the downlink and uplink are 200 Mbps and 100 Mbps respectively. An all fiber network will be more powerful at the heaviest load since it requires the lowest number of nodes per sq. km. While at lighter loads, the IAB deployments perform similarly to an all fiber deployments since the network is less congested and unused resources are leveraged for backhaul.   

The plot of BS density as a function of UE load

IAB can be relied upon for coverage in 5G mmW deployments, and fiber backhaul can be added as the network matures and the 5G UE density increases. In other words, IAB can provide an economically feasible mechanism to quickly provide universal coverage, thus helping to effectively capture more traffic, allowing a straightforward and incremental upgrade pathway to improving capacity over time. Further enhancements to IAB standardization have been planned in future releases of 3GPP standard (Rel-17 and Rel-18) to expand to new use cases and deployments and also to improve the performance in terms of latency and resiliency.

Nokia has been leading the 3GPP standardization effort on IAB since it is an effective tool for cost-effective 5G NR mmWave deployment.

Further reading

[1] Nokia Rel16-17 White Paper:

[2] Amitava Ghosh, “Path towards 5G Millimeter Wave Radio Revolution”, Microwave Journal, June 2016.

[3] Balaz Rel16 3GPP video

Amitabha Ghosh

About Amitabha Ghosh

Amitabha (Amitava) Ghosh (F’15) is a Nokia Fellow and Head, Radio Interface Group at Nokia Bell Labs. He joined Motorola in 1990 after receiving his Ph.D in Electrical Engineering from Southern Methodist University, Dallas. Since joining Motorola he worked on multiple wireless technologies starting from IS-95, cdma-2000, 1xEV-DV/1XTREME, 1xEV-DO, UMTS, HSPA, 802.16e/WiMAX and 3GPP LTE. He has 60 issued patents, has written multiple book chapters and has authored numerous external and internal technical papers. He is currently working on 5G Evolution and 6G technologies. His research interests are in the area of digital communications, signal processing and wireless communications. He is the recipient of 2016 IEEE Stephen O. Rice and 2017 Neal Shephard prize, member of IEEE Access editorial board and co-author of the book titled “Essentials of LTE and LTE-A”.

Mark Cudak

About Mark Cudak

Mark is a Nokia Bell Labs Fellow and a Department Head in the Standardization Research Lab with Nokia Bell Labs in Chicago. His research focuses on cellular network evolution and his team is looking at ways to meet the never-ending growth in wireless data consumption as well as expanding diversity of wireless applications.  Prior to joining Nokia in 2011, Mark was with Motorola for 20 years where he worked on a variety of wireless data systems from 2G to 4G. Mark has over 40 issued patents and holds a M.S. in electrical engineering from the University of Illinois at Urbana-Champaign.

Article tags