Problem & Context
GPS-equipped asset trackers periodically experience “blackouts”, where position data stops updating or becomes unreliable. The events can stem from GPS or cellular loss of signal (LoS). We needed a way to retroactively parse thousands of GPS logs/reports to find these events to prevent them in the future, and make performance improvements of our GPS device.
Scope of Work
- Developed a Python tool that ingests CSV exports of tracker logs and automatically extracts key reliability and performance metrics.
- Expanded the original scope from simply detecting blackout events to computing additional fields that provide a fuller picture of tracker behavior.
- Outputs structured data that can be used directly for debugging, performance analysis, and certification testing.
- Python script developed that could be fed a .csv file of GPS logs/reports to pull data out. The following list is the data that was pulled out of the logs. The scope creeped a little from just collecting the blackout events, but the extra data points really gave a comprehensive understanding of the tracker behavior.
- Data extracted per device/log set:
- 1. Blackout events (GPS and cellular)
- “Go‑dark” events where the device stops reporting indefinitely
- Tracker state (charging vs. off‑charge)
- Time delay from tracker to server (device timestamp vs. server timestamp)
- Cellular carrier/tower changes
- Downtime and distance traveled during blackout periods
Technical Approach
Parsed each log line to extract timestamps, incremental hex counters, GPS coordinates, charge state, and cellular metadata (towers, carriers).
Used the sequential nature of the logs to infer device state:
Repeated GPS timestamps across consecutive reports indicate GPS blackout.
Gaps in the incremental hex counter indicate loss of cellular connectivity.
Applied the Haversine formula to convert latitude/longitude pairs into distances travelled during blackout periods.
Produced a summarized report and machine‑readable output (e.g., CSV) for further analysis and visualization.
This is a snippet of the code…….
Technical diagram or schematic would be displayed here
Results
Project results visualization
Generated per‑device reports listing every blackout and server‑delay event, followed by a concise summary of device performance metrics that can be quickly reviewed by engineers or test teams.
Adapted the tool for GPS certification and regional testing, using the output to visualize blackout‑prone areas and tune device configuration for specific regions to minimize blackouts.