Stacked intelligent metasurfaces (SIMs), composed of a multi-layered structure, have emerged as a powerful computing tool and analog signal processing platform for enabling task-oriented semantic communications (SemCom). However, SIMs lack nonlinear inference capabilities, thus motivating the emergence of the hybrid optical-electronic neural network (HOENN) that cascades a SIM and an electronic neural network (ENN). In this work, we investigate a disaster recognition taskoriented SemCom setting by leveraging the HOENN technology. Specifically, the HOENN is made of an optical neural network (ONN) using SIM mounted on an unmanned aerial vehicle (UAV) and a shallow ENN at the ground receiving station (GRS). The SIM automatically processes semantic information modulated on electromagnetic waves, with reduced energy consumption and ultrafast processing speed. At the GRS, the energy signals are processed by the shallow ENN to enhance the system's inference capability. The aim is to recognize the disaster according to the captured geomorphic images. To this end, we utilize a stochastic gradient descent algorithm to train the HOENN efficiently to minimize the cross-entropy between the recognized and actual semantics. Numerical results show that the HOENN surpasses the performance of using either ONN or ENN alone, achieving$\mathbf{9 7 \%}$recognition accuracy.

Passive wireless human sensing has attracted increasing attention due to its non-contact nature and robustness under different lighting conditions. This paper proposes a system for multiperson detection and vital-sign (i.e., respiration and heartbeat) monitoring using space-time-coding (STC) reconfigurable intelligent surfaces (RISs). Specifically, the proposed system scans the area of interest (AoI) for human detection by using the STC RIS-generated harmonic beams. Simultaneously, specific beams are assigned to the detected persons for vital-sign sensing. Furthermore, we propose an improved variational mode decomposition (VMD) algorithm to process the reflected echo signal and estimate respiration and heartbeat rates of the detected persons. We build a prototype to validate the proposed system. Experimental results demonstrate that the system can simultaneously monitor respiration and heartbeat of up to four persons, with average errors below 1 RPM (respiration per minute) and 5 BPM (beats per minute). Further analysis reveals that the flexible beam control enabled by the STC RIS can reduce the noise reflected by irrelative objects, thus improving the signal-to-noise ratio of echoes from the human chest. This work offers a contact-free care to round-the-clock monitor vital signs of multiple persons, which are critical indicators for assessing general health state and identifying various diseases.