Imageless Shape Detection Using a Millimeter-Wave Dynamic Antenna Array and Noise Illumination

Abstract

We present a dynamic millimeter-wave antenna array with rotational dynamics to detect shapes in a scene without image formation. The rotationally dynamic two-element receiving array is able to capture responses in the spatial Fourier (spatial frequency) domain corresponding to sharp edges in the spatial domain. By implementing a dynamic rotation of 180°, a ring-shaped filter is obtained in the spatial frequency domain. Sharp edges in a scene manifest broad spatial–spectral responses that are localized at specific spatial frequency angles and extend radially outward. The ring filter is orthogonal to these strong responses and thus captures them as a function of spatial frequency angle. Shapes with multiple sharp edges produce responses at angles orthogonal to the edge direction, thereby supporting the ability to classify objects based only on the strength and angular location of the spatial frequency responses. Accurate representation of the spatial Fourier domain requires the scattered fields to be spatially and temporally incoherent, and thus, we transmit noise signals from multiple separate noise transmitters, thereby illuminating the scene with incoherent radiation. The dynamic antenna array thus obtains a high signal-to-noise ratio on the received signals, allowing the use of traditional, low-gain millimeter-wave hardware. We evaluate the concept experimentally with a 38-GHz, two-element dynamic antenna array. We demonstrate the detection of spatial frequency signals via rotational dynamics for several reflecting targets of different shapes. The results demonstrate good agreement with simulations and further support the feasibility of detecting shapes in a scene without image formation.