Mobile operators – especially the visited operator that provides network access to inbound roaming people – find it difficult to monetize inbound M2M / IoT connections due to lack of transparency, limitations imposed by trade agreements, and complexity that surrounds access control for traveling machines. Without transparency and control, operators face competition over connectivity alone, with the inevitable result of commodification of prices and falling market share.
So what can be done to more effectively monetize inbound connections?
It is important to understand that the makeup of roaming subscriptions has changed and will continue to do so due to the ever-increasing number and variety of roaming features. It is no longer reasonable to assume that all homeless are human and can be segmented by traditional criteria such as quiet versus active, business versus casual, and high usage versus low usage.
Today, the fundamental segmentation between human-like communication (HTC) and machine-like communication (MTC) in roaming is crucial. The types of communication exhibit very different characteristics of use and behavior in some scenarios, but very similar characteristics in others.
How do you segment, target and increase roaming revenue, if you can’t tell humans from machines?
Secondly, once the issue of transparency has been resolved, it is essential that, for MTC roaming scenarios, mobile operators are able to collect traffic to their network through a secure and dedicated gateway, allowing operator to also collect M2M traffic. .
The first step focuses on transparency, the second on maintaining control.
There are already a number of solutions available in the market which attempt to meet the challenge of transparency. The vast majority of them rely heavily on Account Transferred Procedures (TAP) or Near Real-Time Roaming Data Exchange (NRTRDE), accounting data and / or device data from International Mobile Equipment Identity (IMEI). This approach applies a set of business rules based on usage patterns or device signatures to classify a rover as a human or a machine.
This classification method for MTC and HTC is no longer appropriate. First, it requires access to Data Clearinghouse (DCH) records such as TAP or NRTRDE streams. Not all roamers generate TAP / NRTRDE records, if there is no usage then there are no usage records. Remember that for most operators a significant percentage of roamers are silent and show no usage.
Another problem is that IMEI data – device data – is not readily available to all roamers. The majority of Signaling System 7 (SS7) transactions, for example, do not have IMEI as a required or optional parameter, which means that roamers that generate only SS7 traffic will not provide IMEI information. Diameter signaling (s6a) and GPRS tunnel protocol (GTP) normally provide the IMEI setting and device information, but this is only true for 60% of roamers, as the remaining 40% generate only roaming SS7 traffic. This means that in the best-case scenario, using DCH and IMEI data to identify human homeless versus machines would only cover around 60% of your overall homelessness, which means the remaining 40% remains unknown.
Finally, to add even more complexity, there are machines that behave like humans and humans that behave like machines from the point of view of use and mobility. Neither usage data nor device data alone can be used to identify features distinctive enough to distinguish the two.
BICS has therefore developed a solution to analyze the raw signaling activity in real time of roaming subscriptions on 2G / 3G / 4G networks and automatically classify the activity of each roamer as MTC or HTC.
Each roamer, whether human or machine, generates signaling traffic data. It doesn’t matter whether they use paid services or not, they have to communicate with the different elements of the network to move around – even silently – and this network level communication occurs more frequently and consistently than any other business activity. roaming.
BICS built a unique machine learning model and used data science to train his model on billions of signaling traffic data records for known machines and known humans. This creates a self-contained classifier capable of identifying MTC and HTC, and thus categorizing homeless people as humans or machines.
This approach can classify 100% of roaming accurately, with 95% accuracy using only SS7 traffic data, and 98% or better accuracy with at least two technologies, such as SS7 and Diameter (s6a) or SS7 and GTP-C or SS7 and GTP-U and Diameter (s6a), as examples.
BICS advanced analyzes combines data science techniques with real-time traffic data and industry-leading data visualization tools to help solve this and other IoT roaming issues. The first version addresses the use cases of inbound M2M and persistent roaming detection, to provide near real-time identification of inbound M2M / IoT subscriptions to help operators monetize existing M2M and IoT connections. on their network.
The system also enables real-time monitoring of known M2M / IoT outbound subscriptions, to monitor and investigate Quality of Service (QoS) issues for deployments and also identify unknown outbound subscriptions to help operators understand and control to what are their SIM cards used for on foreign networks.
Finally, an in-depth analysis of roaming mobility, device distribution and usage profiles is enabled to help operators understand the different behavioral characteristics for each type of connection in their network.
If you would like to know more about BICS Advanced Analytics solution, contact Damion Rose, here.