Impact of Inter-body Scattering on People Counting with Wi-Fi Sensing

We investigate the interactions between human bodies exposed to an incident electric field from a sensing-enabled Wi-Fi access point, to assess how it could affect people counting in the framework of Wi-Fi sensing. We model people with dieletric cylindrical shells with a thickness slightly higher than the electromagnetic skin depth. The electric field scattered by people is computed using the two-dimensional method of moments for electromagnetic scattering, accelerated using an iterative GMRES solver with Adaptive Cross Approximation and a block-Jacobi preconditionner. That scattered field is used to derive channel transfer functions that are combined to obtain a range-Doppler map. The presence of several ghost targets even in simple cases is highlighted, and could hinder people counting. When two people are aligned in front of the sensing-enabled Wi-Fi access point, the first blocking the line-of-sight of the second, it is shown that the field passing through the first person and reflected on the second person is sufficiently strong to detect it. We quantify the corresponding attenuation.