Monad uses WiFi Channel State Information and BLE infrastructure to detect human movement and model crowd dynamics in indoor environments. Exploring how people flowing through spaces create measurable disturbances in wireless signals.
Monad is a Master's thesis research project at the Faculty of Informatics and Information Technologies (FIIT), Slovak University of Technology in Bratislava. The goal is to use Channel State Information (CSI) measurements from deployed WiFi/BLE infrastructure to detect and model human movement in indoor environments.
We have deployed a heterogeneous wireless infrastructure throughout the FIIT library - including BLE beacons, BLE collectors, network communicators, and dedicated CSI monitors operating in AP/monitor mode. A 2D GIS-mapped monitored area with multiple device types captures how people moving through spaces create measurable disturbances in wireless signals.
The core hypothesis is that crowd movement patterns observed through CSI perturbations can be modeled using fluid dynamics analogies - treating crowd flow as a continuum where WiFi signal amplitude and phase changes map to flow properties like velocity and density.
BLE beacons, collectors, CSI monitors, and network communicators are placed across the GIS-mapped monitored area.
Dedicated monitors in AP/monitor mode passively capture WiFi Channel State Information - amplitude and phase across subcarriers.
CSI perturbations caused by human movement are captured alongside BLE RSSI from the mobile app for complementary sensing.
Signal data feeds into fluid dynamics-inspired models to estimate crowd density, flow direction, and movement patterns.
The sensing platform passively collects wireless signal data from multiple device types. No personal data is captured - only RF signal characteristics from the environment.
Channel State Information including amplitude and phase values across subcarriers, capturing how human movement perturbs the wireless signal propagation.
Bluetooth signal strength, device identifiers, and advertisement packets from deployed beacons - complementing CSI for movement detection and positioning.
Millisecond-precision timestamps, GIS-mapped device positions, session metadata, and spatial context linking CSI measurements to physical locations.
Combining CSI-based movement detection with fluid dynamics modeling to understand how crowds behave in indoor spaces.
Channel State Information from WiFi captures signal amplitude and phase changes caused by human movement - enabling passive, device-free sensing.
Exploring fluid dynamics analogies - treating crowd flow as a continuum where CSI perturbations map to flow velocity, density, and direction.
BLE RSSI from deployed beacons and a mobile app provides complementary data for fingerprinting and validating CSI-based detection.
Infrastructure monitored via Consul and Grafana. CSI data, device health, and experiment metrics collected in a unified observability pipeline.
Visit the FIIT STU library in Bratislava and use the Monad Scan app to contribute BLE data. Your movement helps us validate CSI-based crowd models and push the boundaries of passive indoor sensing.
Contact the Researcher