Smart Home IoT Sensor Dataset (1000 rows)

Overview

Synthetic indoor environment sensor data for smart home IoT research. Generated using physics-based models including multi-harmonic Fourier temperature decomposition, CO₂ mass-balance ODEs, and time-dependent Markov occupancy chains.

This dataset contains 1,000 synthetic data points across 8 columns, generated with configuration: Setpoint: 22C.

Column Schema

Mathematical Models

All data points are generated from explicit, documented equations:

• Temperature: Multi-harmonic Fourier: T(t) = T_set + A₁·sin(ω₁t + φ₁) + A₂·sin(2ω₁t + φ₂) + AR(1) noise
• Humidity: Cholesky-correlated with temperature (ρ = −0.65), room-specific modifiers
• CO₂: Mass-balance ODE: dC/dt = (n·G − Q·(C − C_out)) / V
• Occupancy: Time-dependent Markov chain with hourly transition probabilities
• HVAC: Three-state deadband thermostat (±1.5°C hysteresis)
• Light: Dual-source: sin²(solar) + occupancy-driven artificial

Suggested Use Cases

• HVAC optimization
• Occupancy detection/prediction
• Indoor air quality monitoring
• Anomaly detection in sensor networks
• Time-series forecasting benchmarks

Reproducibility

This dataset is fully reproducible. Generation seed: 423138832. Same seed + same parameters = identical output. Generated by IoTSyn v3.1.