Frequency-diverse multimode millimetre-wave constant-ϵr lens-loaded cavity.

This paper presents a physical frequency-diverse multimode lens-loaded cavity, designed and used for the purpose of the direction of arrival (DoA) estimation in millimetre-wave frequency bands for 5G and beyond. The multi-mode mechanism is realized using an electrically-large cavity, generating spatio-temporally incoherent radiation masks leveraging the frequency-diversity principle. It has been shown for the first time that by placing a spherical constant dielectric lens (constant-ϵr) in front of the radiating aperture of the cavity, the spatial incoherence of the radiation modes can be enhanced. The lens-loaded cavity requires only a single lens and output port, making the hardware development much simpler and cost-effective compared to conventional DoA estimators where multiple antennas and receivers are classically required. Using the lens-loaded architecture, an increase of up to 6 dB is achieved in the peak gain of the synthesized quasi-random sampling bases from the frequency-diverse cavity. Despite the fact that the practical frequency-diverse cavity uses a limited subset of quasi-orthogonal modes below the upper bound limit of the number of theoretical modes, it is shown that the proposed lens-loaded cavity is capable of accurate DoA estimation. This is achieved thanks to the sufficient orthogonality of the leveraged modes and to the presence of the spherical constant-ϵr lens which increases the signal-to-noise ratio (SNR) of the received signal. Experimental results are shown to verify the proposed approach.

Large Metasurface Aperture for Millimeter Wave Computational Imaging at the Human-Scale.

We demonstrate a low-profile holographic imaging system at millimeter wavelengths based on an aperture composed of frequency-diverse metasurfaces. Utilizing measurements of spatially-diverse field patterns, diffraction-limited images of human-sized subjects are reconstructed. The system is driven by a single microwave source swept over a band of frequencies (17.5-26.5 GHz) and switched between a collection of transmit and receive metasurface panels. High fidelity image reconstruction requires a precise model for each field pattern generated by the aperture, as well as the manner in which the field scatters from objects in the scene. This constraint makes scaling of computational imaging systems inherently challenging for electrically large, coherent apertures. To meet the demanding requirements, we introduce computational methods and calibration approaches that enable rapid and accurate imaging performance.