We’ve strengthened our anti-fraud system with AI to quickly detect and block SMS fraud. The update also includes message broker integration and automatic removal of numbers from blacklists.
EW SMS Firewall enables SMS aggregators and mobile operators to boost A2P monetization and shield subscribers from spam and fraud. With flexible filtering logic, built-in URL scanning, and AI-powered detection, the system quickly blocks unwanted messages. Advanced analytics provide full transparency and traffic control.
AI Module: Rapid Identification of Complex Fraud Schemes
Challenge
Until recently, managed services specialists manually analyzed SMS traffic to detect suspicious activity, uncover new fraud sources, and adjust filtering rules. This approach had two key limitations:
- Slow response. Analyzing large datasets was time-consuming, which increased the risk of overlooking threats
- Limited coverage. The analysis was based on a narrow sample, lowering the accuracy of conclusions
Solution
We’ve automated full SMS traffic analysis with an AI module to detect complex fraud and spam schemes that are difficult to catch manually. The AI reviews both message content and sender behavior patterns like frequency, destinations, and text.
The AI module is already in use by our clients, helping detect fraud and prevent financial losses from “gray” A2P traffic
The architecture of the AI module is based on an NLP (Natural Language Processing) model that examines both the content and context of each message. This allows it to:
- Detect even well-masked OTP codes
- Identify fraudulent activity in traffic streams
- Predict emerging fraud types, improving accuracy over time and protecting revenue
The AI module is deployed as a supplement to the EW SMS Firewall and supports two operating modes:
- Real-time mode: traffic is analyzed on the fly
- Offline mode: the system scans historical SMS traffic over a defined time range and delivers scheduled reports featuring suspicious numbers and full message texts
Managed services specialists use AI-generated reports to block suspicious numbers and adjust filtering rules accordingly
Apache Kafka Integration: SRI Requests and Fast SMS Processing
Challenge
International A2P SMS is typically delivered via SS7 or direct SMPP links to aggregators, and that’s exactly what EW SMS was originally built for. However, two of our clients—mobile operators in Mozambique and Burundi—have messaging infrastructures based on Apache Kafka.
Solution
To enable integration between EW SMS Firewall and the message broker, we introduced several key updates:
- Integrated an Apache Kafka queue service to manage message flow
- Developed a proxy node to facilitate request and response exchange between the mobile operator and EW SMS Firewall core
- Reworked SMS processing logic by shifting part of the validation from external systems into the core to accelerate processing
- Introduced SRI (Send Routing Information) requests to retrieve recipient routing details in advance and manage delivery cycles
- Upgraded the web interface for improved usability
In our project with a mobile operator in Burundi, we achieved even faster SMS processing cutting response time from 1 second to just 250 milliseconds
“We essentially built a new version of the product to support communication channels it had never handled before. Delivering such a major overhaul under the client’s tight timeline was a real challenge, but we pulled it off and launched right on schedule.”
Andrey Simonov,
Head of The Mobile Messaging Group, Computer Telephony Department, Eastwind
Blacklist Management: Automatic Removal of Numbers
Challenge
Once a number is blacklisted by EW SMS Firewall, the fraudulent user typically stops using it. After 90 days, the mobile operator terminates the contract and makes the number available to new subscribers. If the number stays on the blacklist, new customers may be blocked from accessing mobile services.
Solution
Now, when a number is blacklisted, managed services teams can set an expiration date for automatic removal. They can also track specific numbers and get alerts when the removal period ends making sure new, legitimate users aren’t blocked from using mobile services.
By automatically clearing old numbers from the blacklist, operators can deliver a smoother onboarding experience letting new subscribers use services from day one and building loyalty from the start.
Stat Report Dashboard: Smart Filters for Instant Insights
Challenge
EW SMS Firewall lets clients track stats on every message that’s passed through their network over any time period. Previously, exporting this data required configuring a custom view in the dashboard editor. It was something that demanded an understanding of the database structure, tables, and content. This created friction for clients who don’t work with the system at that level.
Solution
To make reporting easier, we integrated EW SMS Firewall with Metabase OpenSource, a data visualization tool that enables users to build reports, dashboards, and alerts in a convenient, user-friendly interface.
The integration let us upgrade the SMS traffic analytics dashboard with flexible filters. These allow clients to easily customize report views based on their preferred criteria.
Managed services teams handle filter setup for each client, while clients stay in control choosing exactly which fields they want to see in their reports
Additional Enhancements in EW SMS Firewall
Managed services specialists are constantly improving fraud detection rules for faster and more precise blocking. One recent addition is automatic blocking of messages from international senders that mention a brand name. Clients have access to reports detailing the volume and content of these SMS.
Beyond the major updates, we also introduced smaller enhancements throughout the system to make it more efficient and user-friendly.
