Skip to content

user-guide.md

Latest commit

 

History

History
1038 lines (739 loc) · 40.5 KB

user-guide.md

File metadata and controls

1038 lines (739 loc) · 40.5 KB

WiFi DensePose User Guide

WiFi DensePose turns commodity WiFi signals into real-time human pose estimation, vital sign monitoring, and presence detection. This guide walks you through installation, first run, API usage, hardware setup, and model training.

Table of Contents

  1. Prerequisites
  2. Installation
  3. Quick Start
  4. Data Sources
  5. REST API Reference
  6. WebSocket Streaming
  7. Web UI
  8. Vital Sign Detection
  9. CLI Reference
  10. Observatory Visualization
  11. Adaptive Classifier
  12. Training a Model
  13. RVF Model Containers
  14. Hardware Setup
  15. Docker Compose (Multi-Service)
  16. Troubleshooting
  17. FAQ

Prerequisites

Requirement Minimum Recommended
OS Windows 10/11, macOS 10.15, Ubuntu 18.04 Latest stable
RAM 4 GB 8 GB+
Disk 2 GB free 5 GB free
Docker (for Docker path) Docker 20+ Docker 24+
Rust (for source build) 1.70+ 1.85+
Python (for legacy v1) 3.10+ 3.13+

Hardware for live sensing (optional):

Option Cost Capabilities
ESP32-S3 mesh (3-6 boards) ~$54 Full CSI: pose, breathing, heartbeat, presence
Intel 5300 / Atheros AR9580 $50-100 Full CSI with 3x3 MIMO (Linux only)
Any WiFi laptop $0 RSSI-only: coarse presence and motion detection

No hardware? The system runs in simulated mode with synthetic CSI data.

Installation

Docker (Recommended)

The fastest path. No toolchain installation needed.

docker pull ruvnet/wifi-densepose:latest

Multi-architecture image (amd64 + arm64). Works on Intel/AMD and Apple Silicon Macs. Contains the Rust sensing server, Three.js UI, and all signal processing.

From Source (Rust)

git clone https://github.com/ruvnet/RuView.git
cd RuView/rust-port/wifi-densepose-rs

# Build
cargo build --release

# Verify (runs 1,400+ tests)
cargo test --workspace --no-default-features

The compiled binary is at target/release/sensing-server.

From crates.io (Individual Crates)

All 16 crates are published to crates.io at v0.3.0. Add individual crates to your own Rust project:

# Core types and traits
cargo add wifi-densepose-core

# Signal processing (includes RuvSense multistatic sensing)
cargo add wifi-densepose-signal

# Neural network inference
cargo add wifi-densepose-nn

# Mass Casualty Assessment Tool
cargo add wifi-densepose-mat

# ESP32 hardware + TDM protocol + QUIC transport
cargo add wifi-densepose-hardware

# RuVector integration (add --features crv for CRV signal-line protocol)
cargo add wifi-densepose-ruvector --features crv

# WebAssembly bindings
cargo add wifi-densepose-wasm

# WASM edge runtime (lightweight, for embedded/IoT)
cargo add wifi-densepose-wasm-edge

See the full crate list and dependency order in CLAUDE.md.

From Source (Python)

git clone https://github.com/ruvnet/RuView.git
cd RuView

pip install -r requirements.txt
pip install -e .

# Or via PyPI
pip install wifi-densepose
pip install wifi-densepose[gpu]   # GPU acceleration
pip install wifi-densepose[all]   # All optional deps

Guided Installer

An interactive installer that detects your hardware and recommends a profile:

git clone https://github.com/ruvnet/RuView.git
cd RuView
./install.sh

Available profiles: verify, python, rust, browser, iot, docker, field, full.

Non-interactive:

./install.sh --profile rust --yes

Quick Start

30-Second Demo (Docker)

# Pull and run
docker run -p 3000:3000 -p 3001:3001 ruvnet/wifi-densepose:latest

# Open the UI in your browser
# http://localhost:3000

You will see a Three.js visualization with:

  • 3D body skeleton (17 COCO keypoints)
  • Signal amplitude heatmap
  • Phase plot
  • Vital signs panel (breathing + heartbeat)

Verify the System Works

Open a second terminal and test the API:

# Health check
curl http://localhost:3000/health
# Expected: {"status":"ok","source":"simulated","clients":0}

# Latest sensing frame
curl http://localhost:3000/api/v1/sensing/latest

# Vital signs
curl http://localhost:3000/api/v1/vital-signs

# Pose estimation (17 COCO keypoints)
curl http://localhost:3000/api/v1/pose/current

# Server build info
curl http://localhost:3000/api/v1/info

All endpoints return JSON. In simulated mode, data is generated from a deterministic reference signal.

Data Sources

The --source flag controls where CSI data comes from.

Simulated Mode (No Hardware)

Default in Docker. Generates synthetic CSI data exercising the full pipeline.

# Docker
docker run -p 3000:3000 ruvnet/wifi-densepose:latest
# (--source auto is the default; falls back to simulate when no hardware detected)

# From source
./target/release/sensing-server --source simulate --http-port 3000 --ws-port 3001

Windows WiFi (RSSI Only)

Uses netsh wlan to capture RSSI from nearby access points. No special hardware needed. Supports presence detection, motion classification, and coarse breathing rate estimation. No pose estimation (requires CSI).

# From source (Windows only)
./target/release/sensing-server --source wifi --http-port 3000 --ws-port 3001 --tick-ms 500

# Docker (requires --network host on Windows)
docker run --network host ruvnet/wifi-densepose:latest --source wifi --tick-ms 500

Community verified: Tested on Windows 10 (10.0.26200) with Intel Wi-Fi 6 AX201 160MHz, Python 3.14, StormFiber 5 GHz network. All 7 tutorial steps passed with stable RSSI readings at -48 dBm. See Tutorial #36 for the full walkthrough and test results.

Vital signs from RSSI: The sensing server now supports breathing rate estimation from RSSI variance patterns (requires stationary subject near AP) and motion classification with confidence scoring. RSSI-based vital sign detection has lower fidelity than ESP32 CSI — it is best for presence detection and coarse motion classification.

macOS WiFi (RSSI Only)

Uses CoreWLAN via a Swift helper binary. macOS Sonoma 14.4+ redacts real BSSIDs; the adapter generates deterministic synthetic MACs so the multi-BSSID pipeline still works.

# Compile the Swift helper (once)
swiftc -O v1/src/sensing/mac_wifi.swift -o mac_wifi

# Run natively
./target/release/sensing-server --source macos --http-port 3000 --ws-port 3001 --tick-ms 500

See ADR-025 for details.

Linux WiFi (RSSI Only)

Uses iw dev <iface> scan to capture RSSI. Requires CAP_NET_ADMIN (root) for active scans; use scan dump for cached results without root.

# Run natively (requires root for active scanning)
sudo ./target/release/sensing-server --source linux --http-port 3000 --ws-port 3001 --tick-ms 500

ESP32-S3 (Full CSI)

Real Channel State Information at 20 Hz with 56-192 subcarriers. Required for pose estimation, vital signs, and through-wall sensing.

# From source
./target/release/sensing-server --source esp32 --udp-port 5005 --http-port 3000 --ws-port 3001

# Docker
docker run -p 3000:3000 -p 3001:3001 -p 5005:5005/udp ruvnet/wifi-densepose:latest --source esp32

The ESP32 nodes stream binary CSI frames over UDP to port 5005. See Hardware Setup for flashing instructions.

ESP32 Multistatic Mesh (Advanced)

For higher accuracy with through-wall tracking, deploy 3-6 ESP32-S3 nodes in a multistatic mesh configuration. Each node acts as both transmitter and receiver, creating multiple sensing paths through the environment.

# Start the aggregator with multistatic mode
./target/release/sensing-server --source esp32 --udp-port 5005 --http-port 3000 --ws-port 3001

The mesh uses a Time-Division Multiplexing (TDM) protocol so nodes take turns transmitting, avoiding self-interference. Key features:

Feature Description
TDM coordination Nodes cycle through TX/RX slots (configurable guard intervals)
Channel hopping Automatic 2.4/5 GHz band cycling for multiband fusion
QUIC transport TLS 1.3-encrypted streams on aggregator nodes (ADR-032a)
Manual crypto fallback HMAC-SHA256 beacon auth on constrained ESP32-S3 nodes
Attention-weighted fusion Cross-viewpoint attention with geometric diversity bias

See ADR-029 and ADR-032 for the full design.

REST API Reference

Base URL: http://localhost:3000 (Docker) or http://localhost:8080 (binary default).

Method Endpoint Description Example Response
GET /health Server health check {"status":"ok","source":"simulated","clients":0}
GET /api/v1/sensing/latest Latest CSI sensing frame (amplitude, phase, motion) JSON with subcarrier arrays
GET /api/v1/vital-signs Breathing rate + heart rate + confidence {"breathing_bpm":16.2,"heart_bpm":72.1,"confidence":0.87}
GET /api/v1/pose/current 17 COCO keypoints (x, y, z, confidence) Array of 17 joint positions
GET /api/v1/info Server version, build info, uptime JSON metadata
GET /api/v1/bssid Multi-BSSID WiFi registry List of detected access points
GET /api/v1/model/layers Progressive model loading status Layer A/B/C load state
GET /api/v1/model/sona/profiles SONA adaptation profiles List of environment profiles
POST /api/v1/model/sona/activate Activate a SONA profile for a specific room {"profile":"kitchen"}
GET /api/v1/models List available RVF model files {"models":[...],"count":0}
GET /api/v1/models/active Currently loaded model (or null) {"model":null}
POST /api/v1/models/load Load a model by ID {"status":"loaded","model_id":"..."}
POST /api/v1/models/unload Unload the active model {"status":"unloaded"}
DELETE /api/v1/models/:id Delete a model file from disk {"status":"deleted"}
GET /api/v1/models/lora/profiles List LoRA adapter profiles {"profiles":[]}
POST /api/v1/models/lora/activate Activate a LoRA profile {"status":"activated"}
GET /api/v1/recording/list List CSI recording sessions {"recordings":[...],"count":0}
POST /api/v1/recording/start Start recording CSI frames to JSONL {"status":"recording","session_id":"..."}
POST /api/v1/recording/stop Stop the active recording {"status":"stopped","duration_secs":...}
DELETE /api/v1/recording/:id Delete a recording file {"status":"deleted"}
GET /api/v1/train/status Training run status {"phase":"idle"}
POST /api/v1/train/start Start a training run {"status":"started"}
POST /api/v1/train/stop Stop the active training run {"status":"stopped"}
POST /api/v1/adaptive/train Train adaptive classifier from recordings {"success":true,"accuracy":0.85}
GET /api/v1/adaptive/status Adaptive model status and accuracy {"loaded":true,"accuracy":0.85}
POST /api/v1/adaptive/unload Unload adaptive model {"success":true}

Example: Get Vital Signs

curl -s http://localhost:3000/api/v1/vital-signs | python -m json.tool
{
    "breathing_bpm": 16.2,
    "heart_bpm": 72.1,
    "breathing_confidence": 0.87,
    "heart_confidence": 0.63,
    "motion_level": 0.12,
    "timestamp_ms": 1709312400000
}

Example: Get Pose

curl -s http://localhost:3000/api/v1/pose/current | python -m json.tool
{
    "persons": [
        {
            "id": 0,
            "keypoints": [
                {"name": "nose", "x": 0.52, "y": 0.31, "z": 0.0, "confidence": 0.91},
                {"name": "left_eye", "x": 0.54, "y": 0.29, "z": 0.0, "confidence": 0.88}
            ]
        }
    ],
    "frame_id": 1024,
    "timestamp_ms": 1709312400000
}

WebSocket Streaming

Real-time sensing data is available via WebSocket.

URL: ws://localhost:3000/ws/sensing (same port as HTTP — recommended) or ws://localhost:3001/ws/sensing (dedicated WS port).

Note: The /ws/sensing WebSocket endpoint is available on both the HTTP port (3000) and the dedicated WebSocket port (3001/8765). The web UI uses the HTTP port so only one port needs to be exposed. The dedicated WS port remains available for backward compatibility.

Python Example

import asyncio
import websockets
import json

async def stream():
    uri = "ws://localhost:3001/ws/sensing"
    async with websockets.connect(uri) as ws:
        async for message in ws:
            data = json.loads(message)
            persons = data.get("persons", [])
            vitals = data.get("vital_signs", {})
            print(f"Persons: {len(persons)}, "
                  f"Breathing: {vitals.get('breathing_bpm', 'N/A')} BPM")

asyncio.run(stream())

JavaScript Example

const ws = new WebSocket("ws://localhost:3001/ws/sensing");

ws.onmessage = (event) => {
    const data = JSON.parse(event.data);
    console.log("Persons:", data.persons?.length ?? 0);
    console.log("Breathing:", data.vital_signs?.breathing_bpm, "BPM");
};

ws.onerror = (err) => console.error("WebSocket error:", err);

curl (single frame)

# Requires wscat (npm install -g wscat)
wscat -c ws://localhost:3001/ws/sensing

Web UI

The built-in Three.js UI is served at http://localhost:3000/ui/ (Docker) or the configured HTTP port.

Two visualization modes:

Page URL Purpose
Dashboard /ui/index.html Tabbed monitoring dashboard with body model, signal heatmap, phase plot, vital signs
Observatory /ui/observatory.html Immersive 3D room visualization with cinematic lighting and wireframe figures

Dashboard panels:

Panel Description
3D Body View Rotatable wireframe skeleton with 17 COCO keypoints
Signal Heatmap 56 subcarriers color-coded by amplitude
Phase Plot Per-subcarrier phase values over time
Doppler Bars Motion band power indicators
Vital Signs Live breathing rate (BPM) and heart rate (BPM)
Dashboard System stats, throughput, connected WebSocket clients

Both UIs update in real-time via WebSocket and auto-detect the sensing server on the same origin.

Vital Sign Detection

The system extracts breathing rate and heart rate from CSI signal fluctuations using FFT peak detection.

Sign Frequency Band Range Method
Breathing 0.1-0.5 Hz 6-30 BPM Bandpass filter + FFT peak
Heart rate 0.8-2.0 Hz 40-120 BPM Bandpass filter + FFT peak

Requirements:

  • CSI-capable hardware (ESP32-S3 or research NIC) for accurate readings
  • Subject within ~3-5 meters of an access point (up to ~8 m with multistatic mesh)
  • Relatively stationary subject (large movements mask vital sign oscillations)

Signal smoothing: Vital sign estimates pass through a three-stage smoothing pipeline (ADR-048): outlier rejection (±8 BPM HR, ±2 BPM BR per frame), 21-frame trimmed mean, and EMA with α=0.02. This produces stable readings that hold steady for 5-10+ seconds instead of jumping every frame. See Adaptive Classifier for details.

Simulated mode produces synthetic vital sign data for testing.

CLI Reference

The Rust sensing server binary accepts the following flags:

Flag Default Description
--source auto Data source: auto, simulate, wifi, esp32
--http-port 8080 HTTP port for REST API and UI
--ws-port 8765 WebSocket port
--udp-port 5005 UDP port for ESP32 CSI frames
--ui-path (none) Path to UI static files directory
--tick-ms 50 Simulated frame interval (milliseconds)
--benchmark off Run vital sign benchmark (1000 frames) and exit
--train off Train a model from dataset
--dataset (none) Path to dataset directory (MM-Fi or Wi-Pose)
--dataset-type mmfi Dataset format: mmfi or wipose
--epochs 100 Training epochs
--export-rvf (none) Export RVF model container and exit
--save-rvf (none) Save model state to RVF on shutdown
--model (none) Load a trained .rvf model for inference
--load-rvf (none) Load model config from RVF container
--progressive off Enable progressive 3-layer model loading

Common Invocations

# Simulated mode with UI (development)
./target/release/sensing-server --source simulate --http-port 3000 --ws-port 3001 --ui-path ../../ui

# ESP32 hardware mode
./target/release/sensing-server --source esp32 --udp-port 5005

# Windows WiFi RSSI
./target/release/sensing-server --source wifi --tick-ms 500

# Run benchmark
./target/release/sensing-server --benchmark

# Train and export model
./target/release/sensing-server --train --dataset data/ --epochs 100 --save-rvf model.rvf

# Load trained model with progressive loading
./target/release/sensing-server --model model.rvf --progressive

Observatory Visualization

The Observatory is an immersive Three.js visualization that renders WiFi sensing data as a cinematic 3D experience. It features room-scale props, wireframe human figures, WiFi signal animations, and a live data HUD.

URL: http://localhost:3000/ui/observatory.html

Features:

Feature Description
Room scene Furniture, walls, floor with emissive materials and 6-point lighting
Wireframe figures Up to 4 human skeletons with joint pulsation synced to breathing
Signal field Volumetric WiFi wave visualization
Live HUD Heart rate, breathing rate, confidence, RSSI, motion level
Auto-detect Automatically connects to live ESP32 data when sensing server is running
Scenario cycling 6 preset scenarios with smooth transitions (demo mode)

Keyboard shortcuts:

Key Action
1-6 Switch scenario
A Toggle auto-cycle
P Pause/resume
S Open settings
R Reset camera

Live data auto-detect: When served by the sensing server, the Observatory probes /health on the same origin and automatically connects via WebSocket. The HUD badge switches from DEMO to LIVE. No configuration needed.

Adaptive Classifier

The adaptive classifier (ADR-048) learns your environment's specific WiFi signal patterns from labeled recordings. It replaces static threshold-based classification with a trained logistic regression model that uses 15 features (7 server-computed + 8 subcarrier-derived statistics).

Signal Smoothing Pipeline

All CSI-derived metrics pass through a three-stage pipeline before reaching the UI:

Stage What It Does Key Parameters
Adaptive baseline Learns quiet-room noise floor, subtracts drift α=0.003, 50-frame warm-up
EMA + median filter Smooths motion score and vital signs Motion α=0.15; Vitals: 21-frame trimmed mean, α=0.02
Hysteresis debounce Prevents rapid state flickering 4 frames (~0.4s) required for state transition

Vital signs use additional stabilization:

Parameter Value Effect
HR dead-band ±2 BPM Prevents micro-drift
BR dead-band ±0.5 BPM Prevents micro-drift
HR max jump 8 BPM/frame Rejects noise spikes
BR max jump 2 BPM/frame Rejects noise spikes

Recording Training Data

Record labeled CSI sessions while performing distinct activities. Each recording captures full sensing frames (features + raw subcarrier amplitudes) at ~10-25 FPS.

# 1. Record empty room (leave the room for 30 seconds)
curl -X POST http://localhost:3000/api/v1/recording/start \
  -H "Content-Type: application/json" -d '{"id":"train_empty_room"}'
# ... wait 30 seconds ...
curl -X POST http://localhost:3000/api/v1/recording/stop

# 2. Record sitting still (sit near ESP32 for 30 seconds)
curl -X POST http://localhost:3000/api/v1/recording/start \
  -H "Content-Type: application/json" -d '{"id":"train_sitting_still"}'
# ... wait 30 seconds ...
curl -X POST http://localhost:3000/api/v1/recording/stop

# 3. Record walking (walk around the room for 30 seconds)
curl -X POST http://localhost:3000/api/v1/recording/start \
  -H "Content-Type: application/json" -d '{"id":"train_walking"}'
# ... wait 30 seconds ...
curl -X POST http://localhost:3000/api/v1/recording/stop

# 4. Record active movement (jumping jacks, arm waving for 30 seconds)
curl -X POST http://localhost:3000/api/v1/recording/start \
  -H "Content-Type: application/json" -d '{"id":"train_active"}'
# ... wait 30 seconds ...
curl -X POST http://localhost:3000/api/v1/recording/stop

Recordings are saved as JSONL files in data/recordings/. Filenames must start with train_ and contain a class keyword:

Filename pattern Class
*empty* or *absent* absent
*still* or *sitting* present_still
*walking* or *moving* present_moving
*active* or *exercise* active

Training the Model

Train the adaptive classifier from your labeled recordings:

curl -X POST http://localhost:3000/api/v1/adaptive/train

The server trains a multiclass logistic regression on 15 features using mini-batch SGD (200 epochs). Training completes in under 1 second for typical recording sets. The trained model is saved to data/adaptive_model.json and automatically loaded on server restart.

Check model status:

curl http://localhost:3000/api/v1/adaptive/status

Unload the model (revert to threshold-based classification):

curl -X POST http://localhost:3000/api/v1/adaptive/unload

Using the Trained Model

Once trained, the adaptive model runs automatically:

  1. Each CSI frame is classified using the learned weights instead of static thresholds
  2. Model confidence is blended with smoothed threshold confidence (70/30 split)
  3. The model persists across server restarts (loaded from data/adaptive_model.json)

Tips for better accuracy:

  • Record with clearly distinct activities (actually leave the room for "empty")
  • Record 30-60 seconds per activity (more data = better model)
  • Re-record and retrain if you move the ESP32 or rearrange the room
  • The model is environment-specific — retrain when the physical setup changes

Adaptive Classifier API

Method Endpoint Description
POST /api/v1/adaptive/train Train from train_* recordings
GET /api/v1/adaptive/status Model status, accuracy, class stats
POST /api/v1/adaptive/unload Unload model, revert to thresholds
POST /api/v1/recording/start Start recording CSI frames
POST /api/v1/recording/stop Stop recording
GET /api/v1/recording/list List recordings

Training a Model

The training pipeline is implemented in pure Rust (7,832 lines, zero external ML dependencies).

Step 1: Obtain a Dataset

The system supports two public WiFi CSI datasets:

Dataset Source Format Subjects Environments Download
MM-Fi NeurIPS 2023 .npy 40 4 rooms GitHub repo (Google Drive / Baidu links inside)
Wi-Pose Entropy 2023 .mat 12 1 room GitHub repo (Google Drive / Baidu links inside)

Download the dataset files and place them in a data/ directory.

Step 2: Train

# From source
./target/release/sensing-server --train --dataset data/ --dataset-type mmfi --epochs 100 --save-rvf model.rvf

# Via Docker (mount your data directory)
docker run --rm \
  -v $(pwd)/data:/data \
  -v $(pwd)/output:/output \
  ruvnet/wifi-densepose:latest \
  --train --dataset /data --epochs 100 --export-rvf /output/model.rvf

The pipeline runs 10 phases:

  1. Dataset loading (MM-Fi .npy or Wi-Pose .mat)
  2. Hardware normalization (Intel 5300 / Atheros / ESP32 -> canonical 56 subcarriers)
  3. Subcarrier resampling (114->56 or 30->56 via Catmull-Rom interpolation)
  4. Graph transformer construction (17 COCO keypoints, 16 bone edges)
  5. Cross-attention training (CSI features -> body pose)
  6. Domain-adversarial training (MERIDIAN: gradient reversal + virtual domain augmentation)
  7. Composite loss optimization (MSE + CE + UV + temporal + bone + symmetry)
  8. SONA adaptation (micro-LoRA + EWC++)
  9. Sparse inference optimization (hot/cold neuron partitioning)
  10. RVF model packaging

Step 3: Use the Trained Model

./target/release/sensing-server --model model.rvf --progressive --source esp32

Progressive loading enables instant startup (Layer A loads in <5ms with basic inference), with full model loading in the background.

Cross-Environment Adaptation (MERIDIAN)

Models trained in one room typically lose 40-70% accuracy in a new room due to different WiFi multipath patterns. The MERIDIAN system (ADR-027) solves this with a 10-second automatic calibration:

  1. Deploy the trained model in a new room
  2. Collect ~200 unlabeled CSI frames (10 seconds at 20 Hz)
  3. The system automatically generates environment-specific LoRA weights via contrastive test-time training
  4. No labels, no retraining, no user intervention

MERIDIAN components (all pure Rust, +12K parameters):

Component What it does
Hardware Normalizer Resamples any WiFi chipset to canonical 56 subcarriers
Domain Factorizer Separates pose-relevant from room-specific features
Geometry Encoder Encodes AP positions (FiLM conditioning with DeepSets)
Virtual Augmentor Generates synthetic environments for robust training
Rapid Adaptation 10-second unsupervised calibration via contrastive TTT

See ADR-027 for the full design.

CRV Signal-Line Protocol

The CRV (Coordinate Remote Viewing) signal-line protocol (ADR-033) maps a 6-stage cognitive sensing methodology onto WiFi CSI processing. This enables structured anomaly classification and multi-person disambiguation.

Stage CRV Term WiFi Mapping
I Gestalt Detrended autocorrelation → periodicity / chaos / transient classification
II Sensory 6-modality CSI feature encoding (texture, temperature, luminosity, etc.)
III Topology AP mesh topology graph with link quality weights
IV Coherence Phase phasor coherence gate (Accept/PredictOnly/Reject/Recalibrate)
V Interrogation Person-specific signal extraction with targeted subcarrier selection
VI Partition Multi-person partition with cross-room convergence scoring 
# Enable CRV in your Cargo.toml
cargo add wifi-densepose-ruvector --features crv

See ADR-033 for the full design.

RVF Model Containers

The RuVector Format (RVF) packages a trained model into a single self-contained binary file.

Export

./target/release/sensing-server --export-rvf model.rvf

Load

./target/release/sensing-server --model model.rvf --progressive

Contents

An RVF file contains: model weights, HNSW vector index, quantization codebooks, SONA adaptation profiles, Ed25519 training proof, and vital sign filter parameters.

Deployment Targets

Target Quantization Size Load Time
ESP32 / IoT int4 ~0.7 MB <5ms
Mobile / WASM int8 ~6-10 MB ~200-500ms
Field (WiFi-Mat) fp16 ~62 MB ~2s
Server / Cloud f32 ~50+ MB ~3s

Hardware Setup

ESP32-S3 Mesh

A 3-6 node ESP32-S3 mesh provides full CSI at 20 Hz. Total cost: ~$54 for a 3-node setup.

What you need:

  • 3-6x ESP32-S3 development boards (~$8 each)
  • A WiFi router (the CSI source)
  • A computer running the sensing server (aggregator)

Flashing firmware:

Pre-built binaries are available at Releases:

Release What It Includes Tag
v0.2.0 Stable — raw CSI streaming, TDM, channel hopping, QUIC mesh v0.2.0-esp32
v0.3.0-alpha Alpha — adds on-device edge intelligence (ADR-039) v0.3.0-alpha-esp32 
# Flash an ESP32-S3 (requires esptool: pip install esptool)
python -m esptool --chip esp32s3 --port COM7 --baud 460800 \
  write-flash --flash-mode dio --flash-size 4MB \
  0x0 bootloader.bin 0x8000 partition-table.bin 0x10000 esp32-csi-node.bin

Provisioning:

python firmware/esp32-csi-node/provision.py --port COM7 \
  --ssid "YourWiFi" --password "YourPassword" --target-ip 192.168.1.20

Replace 192.168.1.20 with the IP of the machine running the sensing server.

Mesh key provisioning (secure mode):

For multistatic mesh deployments with authenticated beacons (ADR-032), provision a shared mesh key:

python firmware/esp32-csi-node/provision.py --port COM7 \
  --ssid "YourWiFi" --password "YourPassword" --target-ip 192.168.1.20 \
  --mesh-key "$(openssl rand -hex 32)"

All nodes in a mesh must share the same 256-bit mesh key for HMAC-SHA256 beacon authentication. The key is stored in ESP32 NVS flash and zeroed on firmware erase.

TDM slot assignment:

Each node in a multistatic mesh needs a unique TDM slot ID (0-based):

# Node 0 (slot 0) — first transmitter
python firmware/esp32-csi-node/provision.py --port COM7 --tdm-slot 0 --tdm-total 3

# Node 1 (slot 1)
python firmware/esp32-csi-node/provision.py --port COM8 --tdm-slot 1 --tdm-total 3

# Node 2 (slot 2)
python firmware/esp32-csi-node/provision.py --port COM9 --tdm-slot 2 --tdm-total 3

Edge Intelligence (v0.3.0-alpha, ADR-039):

The v0.3.0-alpha firmware adds on-device signal processing that runs directly on the ESP32-S3 — no host PC needed for basic presence and vital signs. Edge processing is disabled by default for full backward compatibility.

Tier What It Does Extra RAM
0 Disabled (default) — streams raw CSI to the aggregator 0 KB
1 Phase unwrapping, running statistics, top-K subcarrier selection, delta compression ~30 KB
2 Everything in Tier 1, plus presence detection, breathing/heart rate, motion scoring, fall detection ~33 KB

Enable via NVS (no reflash needed):

# Enable Tier 2 (full vitals) on an already-flashed node
python firmware/esp32-csi-node/provision.py --port COM7 \
  --ssid "YourWiFi" --password "YourPassword" --target-ip 192.168.1.20 \
  --edge-tier 2

Key NVS settings for edge processing:

NVS Key Default What It Controls
edge_tier 0 Processing tier (0=off, 1=stats, 2=vitals)
pres_thresh 50 Sensitivity for presence detection (lower = more sensitive)
fall_thresh 500 Fall detection threshold (variance spike trigger)
vital_win 300 How many frames of phase history to keep for breathing/HR extraction
vital_int 1000 How often to send a vitals packet, in milliseconds
subk_count 32 Number of best subcarriers to keep (out of 56)

When Tier 2 is active, the node sends a 32-byte vitals packet at 1 Hz (configurable) containing presence state, motion score, breathing BPM, heart rate BPM, confidence values, fall flag, and occupancy estimate. The packet uses magic 0xC5110002 and is sent to the same aggregator IP and port as raw CSI frames.

Binary size: 777 KB (24% free in the 1 MB app partition).

Alpha notice: Vital sign estimation uses heuristic BPM extraction. Accuracy is best with stationary subjects in controlled environments. Not for medical use.

Start the aggregator:

# From source
./target/release/sensing-server --source esp32 --udp-port 5005 --http-port 3000 --ws-port 3001

# Docker
docker run -p 3000:3000 -p 3001:3001 -p 5005:5005/udp ruvnet/wifi-densepose:latest --source esp32

See ADR-018, ADR-029, and Tutorial #34.

Intel 5300 / Atheros NIC

These research NICs provide full CSI on Linux with firmware/driver modifications.

NIC Driver Platform Setup
Intel 5300 iwl-csi Linux Custom firmware, ~$15 used
Atheros AR9580 ath9k patch Linux Kernel patch, ~$20 used

These are advanced setups. See the respective driver documentation for installation.

Docker Compose (Multi-Service)

For production deployments with both Rust and Python services:

cd docker
docker compose up

This starts:

  • Rust sensing server on ports 3000 (HTTP), 3001 (WS), 5005 (UDP)
  • Python legacy server on ports 8080 (HTTP), 8765 (WS)

Troubleshooting

Docker: "no matching manifest for linux/arm64" on macOS

The latest tag supports both amd64 and arm64. Pull the latest image:

docker pull ruvnet/wifi-densepose:latest

If you still see this error, your local Docker may have a stale cached manifest. Try:

docker pull --platform linux/arm64 ruvnet/wifi-densepose:latest

Docker: "Connection refused" on localhost:3000

Make sure you're mapping the ports correctly:

docker run -p 3000:3000 -p 3001:3001 ruvnet/wifi-densepose:latest

The -p 3000:3000 maps host port 3000 to container port 3000.

Docker: No WebSocket data in UI

Add the WebSocket port mapping:

docker run -p 3000:3000 -p 3001:3001 ruvnet/wifi-densepose:latest

ESP32: No data arriving

  1. Verify the ESP32 is connected to the same WiFi network
  2. Check the target IP matches the sensing server machine: python firmware/esp32-csi-node/provision.py --port COM7 --target-ip <YOUR_IP>
  3. Verify UDP port 5005 is not blocked by firewall
  4. Test with: nc -lu 5005 (Linux) or similar UDP listener

Build: Rust compilation errors

Ensure Rust 1.75+ is installed (1.85+ recommended):

rustup update stable
rustc --version

Windows: RSSI mode shows no data

Run the terminal as Administrator (required for netsh wlan access). Verified working on Windows 10 and 11 with Intel AX201 and Intel BE201 adapters.

Vital signs show 0 BPM

  • Vital sign detection requires CSI-capable hardware (ESP32 or research NIC)
  • RSSI-only mode (Windows WiFi) does not have sufficient resolution for vital signs
  • In simulated mode, synthetic vital signs are generated after a few seconds of warm-up
  • With real ESP32 data, vital signs take ~5 seconds to stabilize (smoothing pipeline warm-up)

Vital signs jumping around

The server applies a 3-stage smoothing pipeline (ADR-048). If readings are still unstable:

  • Ensure the subject is relatively still (large movements mask vital sign oscillations)
  • Train the adaptive classifier for your specific environment: curl -X POST http://localhost:3000/api/v1/adaptive/train
  • Check signal quality: curl http://localhost:3000/api/v1/sensing/latest — look for signal_quality > 0.4

Observatory shows DEMO instead of LIVE

  • Verify the sensing server is running: curl http://localhost:3000/health
  • Access Observatory via the server URL: http://localhost:3000/ui/observatory.html (not a file:// URL)
  • Hard refresh with Ctrl+Shift+R to clear cached settings
  • The auto-detect probes /health on the same origin — cross-origin won't work

FAQ

Q: Do I need special hardware to try this? No. Run docker run -p 3000:3000 ruvnet/wifi-densepose:latest and open http://localhost:3000. Simulated mode exercises the full pipeline with synthetic data.

Q: Can consumer WiFi laptops do pose estimation? No. Consumer WiFi exposes only RSSI (one number per access point), not CSI (56+ complex subcarrier values per frame). RSSI supports coarse presence and motion detection. Full pose estimation requires CSI-capable hardware like an ESP32-S3 ($8) or a research NIC.

Q: How accurate is the pose estimation? Accuracy depends on hardware and environment. With a 3-node ESP32 mesh in a single room, the system tracks 17 COCO keypoints. The core algorithm follows the CMU "DensePose From WiFi" paper (arXiv:2301.00250). The MERIDIAN domain generalization system (ADR-027) reduces cross-environment accuracy loss from 40-70% to under 15% via 10-second automatic calibration.

Q: Does it work through walls? Yes. WiFi signals penetrate non-metallic materials (drywall, wood, concrete up to ~30cm). Metal walls/doors significantly attenuate the signal. With a single AP the effective through-wall range is approximately 5 meters. With a 3-6 node multistatic mesh (ADR-029), attention-weighted cross-viewpoint fusion extends the effective range to ~8 meters through standard residential walls.

Q: How many people can it track? Each access point can distinguish ~3-5 people with 56 subcarriers. Multi-AP deployments multiply linearly (e.g., 4 APs cover ~15-20 people). There is no hard software limit; the practical ceiling is signal physics.

Q: Is this privacy-preserving? The system uses WiFi radio signals, not cameras. No images or video are captured or stored. However, it does track human position, movement, and vital signs, which is personal data subject to applicable privacy regulations.

Q: What's the Python vs Rust difference? The Rust implementation (v2) is 810x faster than Python (v1) for the full CSI pipeline. The Docker image is 132 MB vs 569 MB. Rust is the primary and recommended runtime. Python v1 remains available for legacy workflows.

Q: Can I use an ESP8266 instead of ESP32-S3? No. The ESP8266 does not expose WiFi Channel State Information (CSI) through its SDK, has insufficient RAM (~80 KB vs 512 KB), and runs a single-core 80 MHz CPU that cannot handle the signal processing pipeline. The ESP32-S3 is the minimum supported CSI capture device. See Issue #138 for alternatives including using cheap Android TV boxes as aggregation hubs.

Q: Does the Windows WiFi tutorial work on Windows 10? Yes. Community-tested on Windows 10 (build 26200) with an Intel Wi-Fi 6 AX201 160MHz adapter on a 5 GHz network. All 7 tutorial steps passed with Python 3.14. See Issue #36 for full test results.

Q: Can I run the sensing server on an ARM device (Raspberry Pi, TV box)? ARM64 deployment is planned (ADR-046) but not yet available as a pre-built binary. You can cross-compile from source using cross build --release --target aarch64-unknown-linux-gnu -p wifi-densepose-sensing-server if you have the Rust cross-compilation toolchain set up.

Further Reading