Case Studies

Real-time Anomaly Detection Wireless IoT Solutions

Background

A world leader in Telecom Services and Solutions stays ahead of the competition by building innovative solutions and frameworks to help Telecom Service Providers to monetize newer technologies. In that respect, they wanted to build a real time anomaly detection framework based on events from IoT sensors embedded in industrial equipment. They had to build this in weeks to demo it in a summit.

Challenge

Building a framework and an application with a deterministic response time in milliseconds over a short period is nearly impossible. The framework had to support integration to multiple databases, ingestion of large volumes of real-time events and post an instantaneous response to mitigate any risk from an anomaly. On top of the framework will be an application that addresses a specific use case.

Solution

pSolv built the solution using pSolv’s Needletail platform that supports both batch and real-time processing. The Zero Code Needletail platform-based solution ingested real-time events emitted from multiple IoT devices through an MQTT broker(mosquitto), correlated device Id with device information & deployment data provisioned in MongoDB and issued notifications to applications when either temperature or pressure readings were above the configured threshold for a given device type.

The solution was built within two weeks and demonstrated with 600K events per minute with no loss of events. All events were logged to MongoDB and Alerts were published to MQTT and stored in MySQL.

Outcome

  • A versatile framework that can be used to implement multiple use cases
  • Alerts issued within a deterministic time measured in milliseconds
  • A scalable solution that can handle a large event volume

 

Ready to get started?

Fill in your details to experience Needletail in action