Partial dataset (8,774 packets) of indoor LoRaWAN measurements (RSSI, SNR, frequency offset) across residential, laboratory, multi-floor, and underground industrial environments in Casablanca, Morocco. Sample of a 43,100-packet campaign.
# Indoor LoRaWAN Multi-Environment Dataset (Partial Release)
## Overview
This repository contains a **partial sample** of the raw LoRaWAN uplink measurement
dataset collected for the study:
> Aarif, L., Ahaidous, K., Tabaa, M., Hachimi, H., Monteiro, F. \*Empirical
> Characterization and Machine Learning-Based Classification of Indoor LoRaWAN
> Propagation Environments.\*
The full study reports a total of **43,100 packets** collected across four indoor
environments (Table 6 of the paper). **This release contains 8,774 packets**
(≈20% of the full dataset) and is provided as a representative sample to support
transparency and partial reproducibility while the complete dataset is prepared
for a subsequent release. Please contact the corresponding author
(l.aarif@emsi.ma) for access to the complete dataset.
## Files and Environment Mapping
Each CSV file corresponds to one measurement campaign, as described in Section 3.2
of the paper. Files are named ` \_ .csv`, matching the four
environment classes used in the machine learning classification (Section 4.2)
and the class distribution reported in Table 6.
|File|Environment (paper §3.2)|Condition|Rows|Nodes present in this sample|
|-|-|-|-|-|
|`Residential\_LOS.csv`|Residential (apartment)|Line-of-sight|692|Node1–Node3|
|`Residential\_NLOS.csv`|Residential (apartment)|Non-line-of-sight|1,266|Node1–Node9|
|`Laboratory\_LOS.csv`|Research Laboratory|Line-of-sight|718|noeud1–noeud4|
|`Laboratory\_NLOS.csv`|Research Laboratory|Non-line-of-sight|697|noeud1–noeud4|
|`MultiFloorBuilding\_NLOS.csv`|Multi-floor building (EMSI, 5 floors)|Non-line-of-sight (only condition tested)|3,361|noeud1–noeud6|
|`UndergroundIndustrial\_Scenario1.csv`|Underground Industrial (pilot plant)|Scenario 1 (mixed LOS/NLOS)|1,049|noeud1–noeud7|
|`UndergroundIndustrial\_Scenario2.csv`|Underground Industrial (pilot plant)|Scenario 2 (NLOS)|991|noeud1–noeud7|
|**Total**|||**8,774**||
**Note on the four ML classes.** For the supervised classification …