WiFi-CSI Sensing and Bearing Estimation in Multi-Robot Systems: An Open-Source Simulation Framework

TL;DR

This work tackles the hardware bottleneck of WiFi-CSI sensing for multi-robot localization by introducing an open-source Gazebo–Matlab simulation framework that reproduces the WSR toolbox's CSI collection and Bartlett-based bearing estimation. It models carrier frequency offset, sampling time offset, and noise to generate realistic synthetic CSI data, enabling creation of a virtual antenna array through robot motion. The framework is validated against real Turtlebot3 experiments, showing close alignment between CSI phase behavior and bearing estimates in LOS scenarios. By providing a hardware-free development environment, the framework broadens accessibility for designing and testing WiFi-CSI–based multi-robot localization algorithms.

Abstract

Development and testing of multi-robot systems employing wireless signal-based sensing requires access to suitable hardware, such as channel monitoring WiFi transceivers, which can pose significant limitations. The WiFi Sensor for Robotics (WSR) toolbox, introduced by Jadhav et al. in 2022, provides a novel solution by using WiFi Channel State Information (CSI) to compute relative bearing between robots. The toolbox leverages the amplitude and phase of WiFi signals and creates virtual antenna arrays by exploiting the motion of mobile robots, eliminating the need for physical antenna arrays. However, the WSR toolbox's reliance on an obsoleting WiFi transceiver hardware has limited its operability and accessibility, hindering broader application and development of relevant tools. We present an open-source simulation framework that replicates the WSR toolbox's capabilities using Gazebo and Matlab. By simulating WiFi-CSI data collection, our framework emulates the behavior of mobile robots equipped with the WSR toolbox, enabling precise bearing estimation without physical hardware. We validate the framework through experiments with both simulated and real Turtlebot3 robots, showing a close match between the obtained CSI data and the resulting bearing estimates. This work provides a virtual environment for developing and testing WiFi-CSI-based multi-robot localization without relying on physical hardware. All code and experimental setup information are publicly available at https://github.com/BrendanxP/CSI-Simulation-Framework

Figures (6)

• Figure 1: Overview of the operation of our simulation framework (green and orange blocks) in parallel with the WSR toolbox (blue blocks) Jadhav2022ToolboxRobotics. The blocks in green involve implementations in Matlab, Gazebo, and ROS. Those in orange involve Matlab. The WSR blocks in blue involve implementations in C++ and Python Jadhav2022ToolboxRobotics. Three steps are included in our framework: (i) simulating raw CSI and robot odometry data, (ii) computing phase using CSI data, and (iii) running Bartlett's estimator to obtain robots' relative bearing.
• Figure 2: Creating a virtual antenna array by moving a single antenna element on a mobile robot in a SAR-based approach. This example shows a circular array with a 30 cm radius and a transmitting node 4m apart to establish the far-field assumption considered in the WSR toolbox developments.
• Figure 3: We use Turtlebot3 Waffle robots in our simulation and real setups to collect data. (a) Real robot with onboard UP Squared SBC, Intel WL5300 NIC, and WiFi antenna. (b) Simulated robot with the WiFi antenna at the model's center.
• Figure 4: In our experiment setup we use one static robot and one moving robot traversing multiple linear and circular trajectories. The azimuth angle $\theta$ between the two robots is shown, measured with respect to robot $i$'s reference frame.
• Figure 5: Carrier signal phase obtained from both simulation and real experiments after cancellation of CFO. Each data point corresponds to a sample taken at 100 Hz frequency. The real and simulation data match closely for both trajectories.