We shrdy strong coupling (SC) in a system of plasmonic nanohole arrays and transition- metal dichalcogenide material (WS2). Using FDTD simulations and a genetic algorithm, we design several array geometries to obtain large Rabi splitting at room-temperature.

Automotive radar (AR) plays a key role in autonomous vehicles. A modern-day AR requires cognition to adapt to its dynamically changing environment. In an automotive environment, the number of ARs that need to operate simultaneously varies rapidly and these ARs should transceive without mutual interference. Most widely used AR systems use frequency-modulated continuous-wave (FMCW) radar due to their smaller bandwidth and cost compared with pulse Doppler radar (PDR). By using sub-Nyquist based techniques, such as Xampling, we show here that the targets could be estimated from low rate samples even with a PDR. In this paper, we demonstrate a hardware prototype demonstrating the applicability of sub-Nyquist PDR as a cognitive AR. We consider a scenario where a number of ARs are mounted on a vehicle and need to look simultaneously into different directions without interfering with each other. The available bandwidth is divided into several non-overlapping subbands which depend on the number of ARs required. Each AR is assigned a set of randomly spaced subbands for its transceiver to operate. Through simulations, we show that the ARs could detect targets simultaneously without interference. Furthermore, the noise robustness of our sub-Nyquist reconstruction method is better than the standard matched-filtering approach.

We propose a Deep Learning (DL) framework for reconstructing super-resolved images in structured illumination microscopy, which reduces the amount of raw data required for the reconstruction and allows real-time super resolution imaging.