Wi-Fi Based Multi Target Device-Free Sensing Using Joint TX-RX Beamforming

对于F.Y.2021，我们开发了一个能够4x4多输入多输出(MIMO)以及具有2个用户的多用户平台，将用于实施拟议的多目标WiFi传感项目。该平台使用多个ADRV9009 FGPA和一个ZCU102射频子板通过以太网连接，用于板对板通信。使用该平台的传输和接收是通过运行MatLab的主机PC完成的，并连接到FPGA板上的以太网络，利用该平台对其进行实时WiFi侦听的能力进行了测试。目前，测试是使用单一发射设备和单一接收设备进行的。通过这种设置，可以准确地感测到单个WiFi目标。使用MIMO处理，在接收端使用多个天线，测量MIMO信道的特征值的变化，以检测人体的运动。人类的检测是使用由MatLab工具箱提供的传统人工智能辅助系统来完成的。我们的初步结果表明，使用所开发的平台可以很容易地感知到单个WiFi目标。未来，该项目的目标是提供一种改进的系统，允许以与单一目标相似的精度对多个目标进行Wi-Fi传感。

For the F.Y. 2021, we have developed a platform capable of 4x4 Multiple Input Multiple Output (MIMO) as well as Multi-user with 2 users which will be used to implement the proposed Multi-target WiFi sensing project . The platform was built using multiple ADRV9009 FGPA with a ZCU102 RF daughterboards connected via Ethernet for board to board communication. Transmissions and receptions using the platform is done using a host PC running Matlab and connected to the FPGA board ethernet network.Using the platform, we have tested its capability to perform real time WiFi sensing. Currently, the tests were done with a single transmitting device and a single receiving device. With this setup, a single WiFi target is sensed accurately. Using MIMO processing, we used multiple antennas from the receiver and measure the changes in the eigenvalues of the MIMO channel in order to detect human beings movement. The detection of the human beings were done using conventional artificial intelligence aided systems provided by a Matlab Toolbox. Our preliminary results show that single WiFi target can be easily sensed using the developed platform. In the future, the goal of this project is to provide a modified system that allows Wi-Fi sensing of multiple targets with similar accuracy as the single target one.