With Cisco predicting a massive, unprecedented growth in mobile internet traffic in its Visual Networking Index 2016 Report, for the ‘Global Mobile Data Traffic Forecast Update 2010-2015’, it’s clear we’re witnessing a surge in data demand. This exponential increase is driven by the continual rise in the number of smart phones and devices, and an increasing demand for innovative services and unlimited data packages. With the introduction of VoLTE and all-IP networks, capable of delivering volumes of data quickly and efficiently, consumers will be better placed than ever before to benefit from the advantages that new technologies and “all you can eat” data bring.
For telco industry experts, VoLTE is perceived as the next step in the evolution of the third generation partnership project (3GPP). However, when looking to deploy VoLTE networks, there are important fraud considerations that mobile operators must address to ensure they are adequately equipped to deal with the associated risks.
Exponential growth in data traffic brings risk
One of the biggest risks for MNOs is that services grounded in the adoption of VoLTE technology (i.e. delivering superior quality of service (QoS) and high definition (HD) features) are leading to new fraud threats.
Unfortunately for MNOs, VoLTE is even more exposed to fraud than 2G/3G telephony, as signaling is implemented in the mobile OS instead of mobile-based broadband, meaning that vulnerabilities can actually be exploited remotely through mobile malware. Interestingly, most of these vulnerabilities are derived from the SS7 (signaling system version 7) network environment concepts, which forms the set of telephony signaling protocols that were first developed in 1975 to operate the public telephone network (PSTN) telephone calls. These have now been repurposed for VoLTE, and since this technology lives in an all-data-IP-based network, the impact of intrusion may be even greater than in the legacy SS7-based networks.
In order for MNOs to keep profit margins intact and manage massive volumes of data, they therefore need to develop strategies to help them proactively arm their network against fraud attacks, and close the holes left open by the packet-switch networks and all its SS7/SIP fragilities.
Types of fraud in VoLTE, and how to deal with them
Fraudsters are always trying new techniques to bypass MNOs’ controls and gain free access to services. This not only deprives the operator of revenue, but also damages their reputation with subscribers and partners. This is no exception with VoLTE, and fraud management teams must therefore be extremely agile to ensure they’re able to quickly and efficiently respond to a changing fraud landscape. Currently, the most common types of fraud faced in VoLTE are related to charging bypass, call spoofing, mobile malware, and call reselling, IRSF and Wangiri.
To cope with new types of fraud in VoLTE technology, it is vital that MNOs implement faster techniques, such as machine learning. While there are limitations to machine learning, machines are much better than humans at dealing with, and processing, large datasets. They are able to detect and recognise thousands of features on a network event, instead of the few that can be captured by manually creating rules. These systems must be capable of finding patterns of behavior that are different from the normal VoLTE subscriber, and be able to identify suspicious activities and/or fraud attempts, no matter how they are perpetrated, regardless of the scheme used, or the origin/termination point of the fraud.
To address these key capabilities, leading Fraud Management Systems (FMS) nowadays rely on algorithms. These can be designed and tuned to immediately identify the differences between fraudulent and legitimate events, to predict whether an event is likely be fraudulent before it is completed. As a result, a behavioural learning-based FMS is highly beneficial for MNOs, as it dramatically increases their knowledge of fraudsters’ behaviour, and quickly and efficiently generates potential fraud alerts.
For MNOs to support real-time data (voice) services, high volume and best-effort data in the same packet-switched domain, they will be burdened with huge capital and operating expenses from two separate networks. For most operators, VoLTE is the solution to this dilemma.
The volume of services and technologies available to fraudsters makes it a constantly evolving and multi-faceted issue. Actions taken to eliminate fraud modify the fraudsters’ behaviour, leading to ever-changing patterns and renewed threats to operator revenue. In some markets, existing solutions are so ineffective and costly that it is easier to accept the losses than resolve the problem.
This real-time VoLTE environment therefore needs to be supported by innovative and cheaper approaches to effectively protect and monetise not only VoLTE calls, but all customer data traffic. New processes, controls and mechanisms are required for mobile operators to efficiently protect their network against fraud attacks, effectively mitigate risk, prevent revenue leakage and keep customers satisfied. Intelligent controls will enable MNOs’ operation teams to efficiently protect, detect, treat and analyse fraud.
As fraud is becoming more complex and harder to identify, early investigation and detection will be key to helping MNOs tackle fraud faster, sharper and smarter. This will allow MNOs to better protect their network against the increasingly complex fraud schemas brought on by VoLTE.