AI Saves Telcos Money | Can It Grow Revenue | Elisa Industriq
STL Partner's Managing Director of Research, Amy Cameron, on where AI in telecoms stands today: cost savings, revenue growth and trusted, well-governed AI.
Discover how real-time telemetry and AI expose SMS blaster activity and enable rapid fraud detection in complex telecom environments.
Listen to this article
By an experienced network and service assurance team.
Understanding the Challenge of SMS Blaster Detection
Most fraudulent text messages still arrive the boring way: cheap, high-volume SMS sent from overseas through unsuspecting routes. But a more sophisticated threat has been spreading across mobile networks: the SMS blaster.
An SMS blaster is, in effect, a mobile network in a car trunk. It carries its own RAN and core, mimics a legitimate operator, and forces nearby phones onto a fake cell long enough to push fraudulent texts — then hands them back to the real network and drives off. Because the spam never crosses the operator’s SMSCs, traditional content-filtering tools see nothing.
This is the story of how some of Polystar’s long-standing customers turned their core network telemetry into a national defense against this kind of attack.
The Problem: Invisible to Traditional Tools
In this market, the operator runs a complex shared infrastructure with many RAN & Core vendors operated by multiple companies. That fragmentation makes it hard to:
-
Collect consistent radio-side telemetry across all vendors and partners.
-
Build a single national view of mobility anomalies, handovers and failures.
-
Spot devices that broadcast fake cells for less than a minute at a time.
Other markets have tackled SMS blasters using RAN-vendor systems. In a multi-operator, multi-vendor environment, that approach would have meant blind spots and impractical integration.
The Approach: Turn Core-Side Data into Actionable Insights
-
Working with Polystar, they built detections on three signals already flowing through its Osix Monitoring and Kalix Analytics.
-
Spam reports forwarded by subscribers to the global short code, with no matching SMS in network logs — strong evidence the message came from outside the real network.
-
Tracking-area updates that referenced “previous systems” not present anywhere in the operator’s configuration — a sign the device had briefly camped on a fake cell.
-
Handover failures involving non-existent target cells, which cluster geographically wherever a blaster operates.
Because Polystar taps at the S1 interface, the operator captured rich metadata — source cell, target cell, tracking area, identifiers — across all RAN vendors and partners in a single, consistent view. No vendor-specific portal could have done that.
What Mid-Sized Operators Can Take from This
You don’t need a national fraud takedown on your roadmap to draw lessons from this story. The same patterns apply to many of the assurance and customer-experience challenges mid-sized operators face today:
-
Vendor-neutral telemetry matters. In multi-vendor and shared networks, a single, consistent data layer surfaces problems that vendor-specific tools cannot.
-
Core-side data sees more than people expect. Even when traffic never reaches the core, signaling and mobility events expose the underlying behavior.
-
Latency is a business decision. Cutting time-to-insight from minutes to seconds changes what you can do with the data — from reactive reporting to real-time response.
-
Collaboration is part of the platform. The hardest work in this case was joining telecom signals to legal and investigative workflows. Assurance platforms increasingly need to feed downstream teams, not just operations dashboards.
Polystar’s assurance platform — Osix Monitoring for passive probing across 2G to 5G SA, Kalix DataOps for governed data processing, and Kalix Analytics for analytics and AI — was built for exactly this kind of work: turning network data into smart actions, regardless of how complex or multi-vendor the underlying network is.
From Insight to Action in Network Assurance
If you’d like to talk through how this approach applies to your network, we’d be glad to share more case studies and walk through what a similar deployment could look like for your team. Contact Polystar
