SearcharxivSearch

arXiv subjects

Mikhail Tsukerman

Publications and source records attributed to Mikhail Tsukerman.

2 recordsLinked to original sources

Giant Rotational Meta-Doppler from Genetically Designed Superscatterers

The motion of a rigid body interacting with a wave leaves spectral signatures, with the Doppler shift as the dominant contribution. Since any motion can be decomposed into translational and rotational components, rotational Doppler provides additional information about the object's dynamics. In the electromagnetic domain, rotating objects generally produce rotational Doppler, or micro-Doppler, signals determined by the symmetry and spatial structure of the scattering process. For objects that are subwavelength or comparable in size to the wavelength, the response is typically dominated by the lowest dipolar scattering channel, so the leading spectral component commonly appears at twice the angular frequency. Here, we introduce the concept of artificially engineered rotational micro-Doppler by designing a compact, strongly scattering structure that operates through a high-order multipolar cascade of resonances, thereby producing a giant enhancement. Magneto-electric arrays composed of strongly coupled electric and magnetic resonators are optimized in the GHz range using a covariance matrix adaptation genetic algorithm to maximize the micro-Doppler frequency. Unlike conventional higher-order multipole designs used in superscatterers for a specific angle of incidence and polarization, our approach jointly optimizes excitation and scattering under radar-relevant conditions for a rotating blade. The resulting arrays exhibit a giant rotational meta-micro-Doppler response, exceeding the dipolar limit by two orders of magnitude and mapping rotations of tens of hertz into the kilohertz range. Beyond its fundamental significance, this mapping has practical value because it shifts rotor micro-Doppler signatures well above slow-moving radar clutter, thereby improving the detectability of slow motion.

physics.app-ph

Diffusion-Based Electromagnetic Inverse Design of Scattering Structured Media

We present a conditional diffusion model for electromagnetic inverse design that generates structured media geometries directly from target differential scattering cross-section profiles, bypassing expensive iterative optimization. Our 1D U-Net architecture with Feature-wise Linear Modulation learns to map desired angular scattering patterns to 2x2 dielectric sphere structure, naturally handling the non-uniqueness of inverse problems by sampling diverse valid designs. Trained on 11,000 simulated metasurfaces, the model achieves median MPE below 19% on unseen targets (best: 1.39%), outperforming CMA-ES evolutionary optimization while reducing design time from hours to seconds. These results demonstrate that employing diffusion models is promising for advancing electromagnetic inverse design research, potentially enabling rapid exploration of complex metasurface architectures and accelerating the development of next-generation photonic and wireless communication systems. The code is publicly available at https://github.com/mikzuker/inverse_design_metasurface_generation.

cs.LG