Backhaul Aware Mobile Network Management: A Milestone in Reaching Zero-Touch Network Orchestration

Integrated management of access and backhaul networks is critical as mobile operators are trying to cost-effectively manage the rapid rise in mobile traffic and the densification of networks. Its importance will further escalate with the advent of 5G.

Until now, mobile backhaul has been managed as a separate silo with dedicated tools and without using advanced automation techniques. In order to unlock the full potential of mobile networks and deliver distinct 5G services (eMBB, URLLC and mMTC), P.I. Works proposes a holistic approach to network capacity management and quality assurance in which backhaul and access networks are managed together. Early PoCs proved that autonomous configuration and capacity management of both access and backhaul functions together yield significant efficiencies and quality benefits.

Two of the important use cases in 5G network management for integrated access and backhaul network management are: “Backhaul Aware RAN Traffic Steering” and “Dynamic Backhaul Capacity Management”. In both cases, real time monitoring of overutilized and underutilized transport links as well as the prediction of backhaul traffic using AI methods stand out as critical technology assets.

Backhaul Aware RAN Traffic Steering uses AI assisted automation techniques for driving operational efficiency and customer experience. The conventional radio access network load balancing works based on the principal that cells suffering congestion can transfer load to other cells with spare capacity. 

Now, as opposed to traditional load balancing methods that consider RAN performance indicators only (e.g., cell or user throughput, bandwidth utilization), backhaul aware optimization uses transport network data to avoid potential network bottlenecks. AI powered algorithms deliver enhanced QoE by processing the data not only from the radio network, but also from the transport network to improve network capacity utilization and to avoid congestion issues. 

P.I. Works Backhaul Aware RAN Traffic Steering takes the traditional load balancing to the next level and looks into the availability of the transport link capacity of candidate cells before a hand over action is taken. Hand over to a cell with limited backhaul capacity will not solve congestion problems. The AI assisted predictive approach is also an important capability of the solution. With this capability, P.I. Works solution can look into the existing capacity of the candidate cell’s transport link as well as the predicted utilization level of that link. 

AI Assisted Dynamic Backhaul Capacity Management solution analyzes the bandwidth utilization of each transport link connecting radio access network to core network. As in Backhaul Aware RAN Traffic Steering, the solution uses predictive algorithms to forecast the utilization of the backhaul capacity after extracting the seasonal, deseasonal, and trend data on each transport and radio link. And the adaptive and configurable thresholding mechanism detects underutilized and overutilized transport links based on the current and predicted backhaul network traffic. Then the capacity of transport links is either downgraded or upgraded based on the current and future capacity requirement. 

In a recent trial with a European Tier 1 Mobile Operator with approximately 20 million subscribers, P.I. Works demonstrated that backhaul aware optimization offers significant cost benefits by reducing both OpEx and CapEx and ensures service quality. The solution recommended bandwidth upgrade on 6 leased transport links and bandwidth downgrade on 266 leased transport links. In the PoC area total savings of $0.16 million / year were achieved through centralized management of the backhaul network. This corresponds to estimated savings of $1.5 million / year in case of network-wide deployment. This PoC is also an important milestone in the evolution towards Zero-Touch Network Operations as mobile network automation is moving beyond the radio access network. 

Many operators are realizing the critical importance of fiber transport in 5G deployment. John Donovan, CEO of AT&T Communications, recently stated that “A lot of the business cases on 5G is going to be how dense is your fiber network. The bigger expense tends to be monthly OpEx on the fiber, the hanging locations and the rights of way.” Tier 1 operators are already signing backhaul deals with Cable operators; and leading operators, like Orange, are investigating more efficient backhaul solutions to support 5G RAN architecture.

We see that the automated management of backhaul networks is becoming a necessity and integrated mobile network management will be the de facto industry standard on the way to 5G. P.I. Works will continue to drive innovation in AI assisted network automation and backhaul will be an integral part of end to end 5G network management. For more information please contact

About P.I. Works

P.I. Works is the leading provider of AI-driven mobile network planning, management and optimization solutions. P.I. Works combines field-proven expertise with its award-winning product portfolio and services. These solutions empower Mobile Operators to accelerate network transformation, improve network quality and reduce network management costs on the way to 5G. P.I. Works has deployed its solutions at 70 mobile operators across 48 countries. P.I. Works also plays an important role in the development of key standards that define the future of mobile networks. We actively contribute to the ETSI, GTI, 3GPP standardization forum and Open Source initiatives.


P.I. Works - Istanbul Teknopark,
Sanayi Mah. Teknopark Bul.
No:1/9C 34906 Pendik/Istanbul
T: + 90 216 265 1123
F: + 90 216 912 1099

We use cookies to improve your user experience, deliver content tailored to your interests and display advertisements through the collection of your IP Address and related browsing statistics. When you use our website, you will be deemed to have accepted the collection of your personal data by cookies. To learn more, please read our privacy policy.