Step 4: Capture and detect an event in the data with a Rule

Overview

Now that you have explored the dataset and visualized key signals, this step shows you how to formalize your observation by creating a Rule. In Sift, Rules are reusable logic expressions that automatically detect patterns or thresholds in telemetry data. When you notice an unusual behavior such as a sudden temperature drop, you can turn that insight into a Rule so that future data is automatically monitored for similar conditions.

In the dataset, we can see a sharp and sudden decline in both max_air_temp(C) and max_ground_temp(C). The image below shows that on October 6, 2021, at 17:00:00, both signals dropped well below their normal range. This event may reflect an unexpected environmental condition, like a cold front, or possibly a brief telemetry issue. By capturing this behavior in a Rule, you ensure that future datasets are automatically checked for similar anomalies and brought to your attention for review.

Create a Rule

To monitor this anomaly in future data, we will define it as a Rule that Sift can automatically detect and highlight during ongoing telemetry analysis.

1. In Sift, click .
2. Click the Rules tab.
3. Click New Rule.
4. In the Rule name box, enter the following:

   fl_low_max_temp_anomaly

   Replace fl with your own initials.

5. In the Asset name, select your Asset (for example, fl_mars_env_monitoring_station).
6. In the Input Channels list, select the following Channels:
   • max_air_temp(C)
   • max_ground_temp(C)
7. In the Enter an expression box, enter:

   $1 < -50 && $2 < -60

8. Click Save.