Kloud Insights

The Kloudspot Analytics Platform provides a high-speed scalable stream processing focused on analysis streams of BLE/WiFi client location and presence data, IoT sensors including CO2, humidity, temperature, in combination with other data sources such as guest portal registrations.

The main flow is as follows:

• The Event Receivers take in event data from the configured Kloudspot or third-party data sources. The received events are transformed from the, normally, proprietary formats used to a consistent internal format that is then sent to the stream processing Job.
• The Stream Processing Job is where all the high-performance processing in our system is performed. We use Apache Flink due to its ability to scale to handle massive loads. The job does the following main things:
• Convert the events into a set of ‘sessions’ - one per client.
• Locate each client in real-time using one of several techniques:
• Presence, where the client is located in a zone based on single strength to nearby Access Points.
• Trilateration, where multiple signal strength measurements are used to locate a client to a specific X/Y or geo coordinate.
• Aggregate the client information to provide time series counts based on various attributes.
• Perform path-movement analysis on the clients.
• Look for matches against the rules.
• The main application provides
• Flexible data presentation.
• Rule creation and editing.
• Sending notifications (for example SMS, or Email).
• Scheduled and on-demand reporting.
• Configuration and monitoring of the system.
• Application programming interface (API)

In this documentation