In our last blog post, we discussed the widespread offerings of data-centric services, the increased focus on Quality of Service (QoS), and how it needs to be tailored to the different customer segments, consumers, enterprises, and IoT providers. We presented the concept of Adaptive Steering, which adjusts steering decisions depending on the scoring of partners’ network QoS and subscriber segments.

Suitable for Individual Purposes

Before jumping into the details, it’s important to reinforce the need for a personalized steering logic beyond network coverage or technology. This is not a new concept, but something we have dealt with for a few years. Traditional discount agreements between roaming partners for basic voice, data, and SMS targeting the retail consumer segment were often translated into dedicated steering policies for prepaid, postpaid, or corporate. These policies had different enforcement levels, some favoring wholesale cost reduction, and others favoring Quality of Experience (QoE). VoLTE plays a fundamental role in evolving voice communication, however, its dependency on handset models and firmware forces Steering of Roaming (SoR) services to apply an extra logic on top of any pre-defined segments to ensure continuous and seamless voice services. Similarly, when exploring modern connectivity services, operators have to deal with cases where a group of industrial sensors from company A may be connected to a private network with an assured QoS and another family of sensors from company B, located in the same place, but connected to a public 5G SA network slice. Therefore, advanced SoR must have a holistic understanding of the entire roaming ecosystem, customer base, connectivity requirements, and commercial objectives.

Automation Is Everything

One of our major drivers is to promote automated roaming services, allowing a smooth flow of data collection, analysis, and actions with reduced manual interruptions, to avoid human errors and offer operational efficiencies for the operations teams. The Adaptive Steering feature of Intelligent Preferred Network (IPN) was built with this premise. For data collection, a GPRS Tunneling Protocol (GTP) probe is used to monitor data session activation, update, deactivation, and QoS parameters. Here are some problematic scenarios considered:

• Network quality: High failure rates or no data session created.
• Coverage: Coverage loss during handover.
• Handovers: Too many handovers may indicate a drop in QoS.
• Session duration: Short average data session duration may also indicate low QoS.
• Data usage: No data roaming usage/low usage scenarios.
• Throughput: Low network data throughput.

Metrics can be measured in different granularities, i.e., at subscriber, APN, and location level. All this information is compiled as part of the Intelligent Preferred Network (IPN) logic to attribute a QoS scoring to partners’ networks and classify subscribers into dynamic communities with special steering policies. This scoring is refreshed periodically to ensure data is reliable. The decision-making process is based on operator-defined targets. In principle, if the scoring is below the target threshold, IPN avoids new registrations in said network, otherwise, the service does not enforce any special policy and the steering decision is driven as defined in the configuration by the operator.

Adaptive Steering can also be extended to measure subscribers’ data sessions aiming to apply roaming usage controls, which eliminates high usage charges for specific segments. This is accomplished through blocking and unblocking actions in real-time.

The expected result is an improved user experience for consumers and enterprise customers with roaming costs optimized for the operator in an increasingly challenging roaming ecosystem. IPN Adaptive Steering also offers comprehensive analytics on networks’ QoS and devices, which can further be used to ensure that any QoS business model is respected. 5G Standalone (SA) may be the precursor of such initiatives, which allow operators to improve their revenues with differentiated offerings, especially in the enterprise IoT market.

Don’t hesitate to contact us to learn more about IPN Adaptive Steering, and how it can be used to overcome the challenges of QoS in roaming services today.