LX-Beats

Taking live heart-rate from 60 people at the same time using 20 NPG Lite devices connected to a single ESP32-S3 dongle to control a live animation on a web interface!

Data Structure:

• 20 nodes × 3 persons each = 60 simultaneous heart rate readings
• RMS calculation aggregates BPM from all connected nodes to a single BPM value
• Time-series data with event markers for performance analysis

Setup

1. Place all files in the same directory
2. Serve via local web server: python -m http.server 8000
3. Open http://localhost:8000/index.html

Features

• BPM Plot: Color-coded heart rate bars (Green <90, Orange 90-110, Red >110)
• Nodes Plot: Device connectivity (Red <5, Orange 5-10, Green >10 connected devices)
• Event Markers: Purple circles mark performance events - hover for full data
• Performance Ranges: Red background indicates active performance periods
• Interactive Timeline: Drag, zoom, and scroll through time-series data
• Dual Data Sources: Toggle between Raw and Processed data instantly

Navigation & Controls

Timeline Navigation:

• Drag Timeline Bar: Click and drag the grey timeline bar to scroll through time
• Mouse Wheel: Scroll up/down to move forward/backward through data
• Zoom Slider: Drag left/right arrows on the slider to adjust visible time window
• Slider Handle: Drag the entire green slider handle to jump to different time periods

Data Interpretation:

• BPM Bars: Green (<90) = relaxed, Orange (90-110) = normal, Red (>110 or clamped) = elevated
• Nodes Bars: Red (<5) = poor connection, Orange (5-10) = moderate, Green (>10) = good connection
• Purple Markers: Event indicators (hover for details) - clickable circles at plot tops
• Red Background: Performance duration (from start to freeze frame)
• Hover Any Bar: Shows Time, BPM, Connected Nodes, and Confidence %

Data Source Toggle:

• Switch between "Raw Data" and "Processed Data" buttons at top-right
• Data is cached for instant switching

Data Processing

The project includes a Jupyter notebook ISF-2026-Data-Processing.ipynb for processing raw BPM data:

Processing Steps:

1. Event Marker Integration: Merges event markers from ISF-2026-Event-Markers.csv into the main dataset
2. Raw Data Export: Generates ISF-2026-BPM-Data-Raw-with-events.csv with all 20 nodes
3. Data Filtering: Creates processed dataset by:
   • Removing Node14 (disconnected device)
   • Filtering out rows with ConnectedNodes < 2
4. Processed Data Export: Generates ISF-2026-BPM-Data-Processed-with-events.csv

Requirements: pandas, pathlib

Usage: Run all cells sequentially to regenerate both CSV files from source data.

Files Required

• ISF-2026-BPM-Data-Raw-with-events.csv
• ISF-2026-BPM-Data-Processed-with-events.csv
• index.html

---