Radio device anomaly detection

A method of identifying anomalies within radio devices, the method comprising:

• at a data analytics platform and via an electronic processor of the data analytics platform,
• obtaining raw radio data from a radio device of a plurality of radio devices assigned to a group within a predefined geographic area;
• identifying, from the raw radio data, a radio performance indicator related to an operational performance of the radio device, the radio performance indicator being associated with a radio communication frequency of the group;
• determining a long-term average and a short-term average of the radio performance indicator based on previously received radio performance indicators from the radio device;
• determining a weighted ratio of the long-term average and the short-term average based on an age of the radio performance indicator and an amount of the previously received radio performance indicators from the radio device;
• generating a curated radio performance indicator from the weighted ratio and the radio performance indicator;
• deriving, from a plurality of curated radio performance indicators from the plurality of radio devices and the curated radio performance indicator, a curated group average of the radio performance indicator of the group;
• determining an anomaly in the operational performance of the radio device based on a comparison of the curated radio performance indicator and the curated group average of the radio performance indicator;
• determining a recommended action to address the anomaly;
• and
• transmitting an electronic notification indicating the recommended action.