How WiFi CSI Sensing Works
An accessible walkthrough of WiFi Channel State Information sensing — from raw packets to AI detection.
WiFi CSI sensing represents a giant leap forward from standard RSSI metrics. While RSSI only records the general signal power of a received packet, **Channel State Information (CSI)** records the complex numeric amplitude and phase shifts of every individual OFDM subcarrier. As these subcarrier frequencies travel through a room, they bounce off solid surfaces and human bodies. Because human skin is highly dielectric due to water content, walking or moving through the room creates a unique, scrolling electromagnetic interference spectrogram. This spectrogram is then analyzed in real-time by edge neural networks to classify activities.
Why RSSI is Insufficient for Spatial Intelligence
RSSI is a single, highly unstable scalar value. If a door swings open or a microwave turns on, RSSI fluctuates wildly without indicating *why*. CSI, however, acts as a high-resolution 3D grid. By monitoring 56 or 114 separate subcarrier channels simultaneously at 100 Hz, we can map exact wave patterns and isolate breathing patterns from standard ambient noise.
Figure 1: High-fidelity conceptual render analyzing How WiFi CSI Sensing Works.
Electromagnetic Wave Propagation & CSI Physics
To fully grasp how wireless sensing works, we must investigate the mathematical principles of modern radio frequency (RF) propagation. Traditional signals like RSSI only provide the average overall power of a received wireless packet. Conversely, Channel State Information (CSI) extracts complex vectors mapping individual Orthogonal Frequency-Division Multiplexing (OFDM) subcarrier channels. In a standard 20 MHz or 40 MHz WiFi spectrum, the signal is split into 56 to 114 separate subcarrier channels. For each subcarrier, the CSI packet header records the exact Amplitude (signal attenuation) and Phase (fractional cycle shift).
Human bodies are comprised of more than 60% water, making them highly conductive dielectric objects in the path of 2.4 GHz and 5.8 GHz frequencies. As waves travel between the transmitter and receiver, they bounce off walls, obstacles, and humans in a phenomenon known as Multipath Propagation. The physical displacement of a human body perturbs this multipath beam network, creating constructive and destructive interference waves. For a comprehensive overview of how these physical shifts are visualized in real-time, try our Interactive 3D WiFi Radar Demo.
Digital Signal Processing & CSI Denoising Pipelines
Raw CSI data streamed from low-cost IoT chips is inherently messy. Thermal drift, clock phase offsets, and ambient environmental interference introduce massive high-frequency noise. The first step in our digital signal processing (DSP) pipeline is running the Base64 stream through a Hampel outlier filter.
We then apply Principal Component Analysis (PCA). PCA analyzes the covariance matrix across all 56 subcarriers and extracts the top three dominant variance components, effectively discarding redundant channels. To monitor rhythmic body signals such as respiration, the clean PCA streams are passed through a Butterworth bandpass filter tuned between 0.15 Hz and 0.35 Hz. For a detailed guide on how we isolate chest-wall displacements using these filters, proceed to Breathing & Vital Signs Detection.
Figure 2: Technological block diagram demonstrating Digital Signal Processing & CSI Denoising Pipelines.
Deep Learning Architectures for Human Activity Recognition
Once raw radio signals are clean, they are formatted as a 2D spectrogram (time vs. subcarrier amplitude values). This allows us to apply advanced Computer Vision algorithms. A 2D Convolutional Neural Network (CNN), such as a modified ResNet, scans the spectrogram to detect distinct 'micro-Doppler' signatures.
To capture temporal actions (such as tracking if a person is sitting down slowly or falling down suddenly), we feed the spatial CNN features into a Long Short-Term Memory (LSTM) recurrent network. On localized edge servers (like a Raspberry Pi 4), this hybrid CNN-LSTM pipeline runs in under 25ms with 96% classification accuracy. You can read a complete architectural comparison of these AI models versus traditional security lenses in our guide WiFi Radar vs Surveillance Cameras.
FAQ
How is How WiFi CSI Sensing Works implemented practically?
Implementing this practically involves deploying inexpensive ESP32 microcontrollers, flashing standard CSI firmware, and routing the Base64 serial streams to a local edge processor running the RuView AI model array.
Does this setup interfere with my home WiFi?
No. The system uses standard IEEE 802.11 beacons and passive sniffing modes, operating seamlessly in the background without affecting your network speed or causing interference.