Wi-ESP (ESP based CSI Tool)

1. GitHub
2. Paper
3. Documentation
4. Download
5. KIST-WRL
6. Datasets

Device-Free Wi-Fi Sensing (DFWS) has garnered significant attention in research circles, thanks to the widespread availability of Wi-Fi signals and its capability to provide relatively precise sensing with minimal infrastructure requirements. One of the primary advantages of DFWS lies in its ability to harness the omnipresent Wi-Fi signals for passive detection of various human activities such as movement, falls, gestures, and location tracking, among others. Unlike traditional sensing methods that rely on dedicated devices like sensors or cameras, DFWS utilizes the inherent patterns within Wi-Fi signals to discern human presence and behavior. This technology holds promise across diverse domains such as the Internet of Things (IoT), Human Activity Recognition (HAR), and elderly care, owing to its unobtrusive nature and ability to facilitate pervasive sensing. The inception of 'Sensorless Sensing' can be traced back to early efforts in human detection utilizing Wireless Sensor Networks (WSN) and the Received Signal Strength Indicator (RSSI). These pioneering endeavors laid the groundwork for subsequent advancements in localization and tracking, paving the way for the widespread adoption of DFWS-based applications. A notable evolution in DFWS came with the utilization of Channel State Information (CSI) in the 802.11n standard, offering a finer-grained insight into the Physical (PHY) layer. This enhancement enabled the extraction of valuable information at the frame level, augmenting the capabilities of DFWS beyond what was achievable with RSSI-based methods."

Download Documentation and Toolkit

Publications

1. Wi-ESP: Wi-ESP—A tool for CSI-based Device-Free Wi-Fi Sensing (DFWS)
2. COVID-Beat: a low-cost breath monitoring approach for people in quarantine during the pandemic
3. HAR: Device Free Wireless Sensing based Human Activity Recognition Using Commercial Off-the-Shelf IoT Single-Board Computers

Public Demonstrations

1. HAR at Living Lab
2. Overall concept
3. CES-2024

Citations

Wi-ESP around the globe

1. Real-Time Operating System for Multitasking Control in the Robotics and Automation Industry
2. One-Class Support Vector Machine for WiFi-based Device-free Indoor Presence Detection
3. WiFi Sensing Model for Intrusion Detection in Smart Home Environment
4. IoT Enabled Programmable SwitchBox with existing SwitchBoard back
5. Distance Estimation Between Wireless Sensor Nodes Using RSSI and CSI with Bounded-Error Estimation and Theory of Evidence for a Landslide Monitoring System
6. Electromagnetic Models for Device-Free Radio Localization with Antenna Arrays
7. WIP: Impulsive Noise Source Recognition with OFDM-WiFi Signals Based on Channel State Information Using Machine Learning
8. STrack: Velocity Estimation Using Single Antenna WiFi Devices
9. Activity Recognition: Device Free Wireless Sensing based Human Activity Recognition Using Commercial Off-the-Shelf IoT Single-Board Computers
10. An EM Body Model for Device-Free Localization with Multiple Antenna Receivers: A First Study
11. Presence Detection with Wi-Fi Using ESP32
12. Non-Contact Wi-Fi Sensing of Respiration Rate for Older Adults in Care: A Validity and Repeatability Study
13. A Low Cost Modular Radio Tomography System for Bicycle and Vehicle Detection and Classification
14. On the Impact of the Antenna Radiation Patterns in Passive Radio Sensing
15. Collecting Channel State Information in Wi-Fi Access Points for IoT Forensics
16. Contactless WiFi Sensing and Monitoring for Future Healthcare - Emerging Trends, Challenges, and Opportunities
17. Device-free Pedestrian Count Estimation Using Wi-Fi Channel State Information
18. WiFi-Based Cross-Domain Gesture Recognition via Modified Prototypical Networks
19. Deep Learning-Based Fall Detection Using WiFi Channel State Information
20. Decentralized ME-Centric Framework—A Futuristic Architecture for Consumer IoT
21. Wi-Sense: a passive human activity recognition system using Wi-Fi and convolutional neural network and its integration in health information systems
22. Wi-SafeHome: WiFi Sensing Based Suspicious Activity Detection for Safe Home Environment
23. WiFi-Based Multi-task Sensing
24. Wi-Monitor: Wi-Fi Channel State Information-Based Crowd Counting with Lightweight and Low-Cost IoT Devices
25. Touchless Biometric User Authentication Using ESP32 WiFi Module
26. WiFi CSI-Based Long-Range Through-Wall Human Activity Recognition with the ESP32
27. Wi-Fi Signal-Based Through-Wall Sensing for Human Presence and Fall Detection Using ESP32 Module
28. Internet of Things
29. Mode confusion of human machine interfaces for automated vehicles
30. CSI-based human sensing using model-based approaches: a survey
31. COVID-Beat: a low-cost breath monitoring approach for people in quarantine during the pandemic
32. A Channel State Information based Respiration Rate Monitoring Method
33. WiFi Sensing with Single-Antenna Devices for Ambient Assisted Living
34. A Platform for Inpatient Safety Management Based on IoT Technology
35. Car-Sense: Vehicle Occupant Legacy Hazard Detection Method Based on DFWS
36. Addressing IoT storage constraints: A hybrid architecture for decentralized data storage and centralized management
37. A survey on vital signs monitoring based on Wi-Fi CSI data
38. Implementation of a wireless text data transmission based on the impulsive control of fractional-order chaotic systems
39. Motion Detection using CSI from Raspberry Pi 4
40. Directional Antenna Systems for Long-Range Through-Wall Human Activity Recognition
41. Channel state information-based fall detection using IoT devices
42. Development of a Non-Invasive Blood Glucose Monitoring Device Using Machine Learning Technology
43. AI enabled RF sensing of Diversified Human-Centric Monitoring
44. A Platform for Inpatient Safety Management Based on IoT Technology
45. End-to-end security enabled intelligent remote IoT monitoring system
46. Artificial intelligence driven Wi-Fi CSI data mining: Focusing on the intrusion detection applications
47. A comprehensive review of wifi sensing technologies: Tools, challenges and future research directions
48. Machine Learning for Wireless Networks - Recent Advances and Future Trends

Datasets

1. Coming Soon......

Other CSI Acquistion Tools

1. Coming Soon......

Wi-ESP is Why-ESP?

ESP32 is a low cost IoT device can also be used to acquire CSI. This provides low cost deployment of the solution to the problem.

• Provide Amplitude and Phase information of the CSI.
• Not limited to LLTF but also HT-LLTF and other parameters.
• Can be used an edge device for computation.
• Can be used for application development.
• Reconfigurable according to application.

Steps to acquire CSI by ESP32

This toolkit is used with WROOM 32 and ESP32 WROVER

Steps to acquire CSI are as follows:
• Step-1: Setup the wireless network to acquire CSI.
• Step-2: Make Client/Server architecture to generate traffic.
• Step-3: Acquire CSI to configure ESP32 on the same channle on which Wireless Network is established.

Earlier Version of Espressif

ESP32 console based Configuration Settings

There are few configurations which are very necessary to acquire CSI. We have included all these settings in the manuscript and discuused very much in detail. These setting includes CPU clock setting, we need to set at maximum so that we can acquire CSI at maximum rate. UART baud rate is very necessary to acquire CSI from ESP32 board. This settings depends on the board type of ESP32 we are using to acquire CSI. Third settings is the SSID information, which need to set before acquring the CSI. The last but the most important setting is the Enabling to get CSI. All these settings can be performed thorigh source code but the more easy way is by using console.

There are different options are available to set the clock frequency of the processor. The best is to select maximum which is 240MHz. In some cases this is observed that low clock rate can also work but this is better to select the clock rate maximum.

The baud rate settings is necessary in case you are using UART port to commuincate to the ESP32. The Baud rate depends upon the board type you are using to acquire CSI. There are many kinds of board are available in the market. Select that board which provides you maximum baud rate which is 2M.

The CSI gathering can be performed without using this. But still we have done this to avoid any confusion. The important thing to acquire CSI is the channel number which are you uisng for transmission This is better to filter the channel number and MAC address to minimze the traffic otherwise the traffic need to be filter at later stage.

The most important setting of the ESP32 is the enabling the CSI by using the console. This must be enable before configuring ESP32.

Reach Us

Tool developed by: Dr. Muhammad Atif | WRL Lab @ KIST South Korea

Cite this tool: BibTeX