Study of Terahertz Superresolution Imaging Scheme With Real-Time Capability Based on Frequency Scanning Antenna

In this paper, a terahertz (THz) imaging scheme with real-time capability is proposed based on frequency- controlled beam scanning. The scheme holds a concise system architecture comprising a frequency scanning reflector antenna (FSRA) as the transmitting antenna, an omnidirectional antenna as the receiving antenna, and a vector network analyzer as the transceiver. By employing the diffraction enhancement mechanism to improve the diffraction efficiency and to suppress the specular beam, a high-directivity FSRA with periodic planar binary structure is proposed and developed at 0.235–0.33 THz band for imaging applications. A superresolution algorithm is proposed and developed by combining the direction of arrival (DoA) estimation and the matched filter technique, in which the tradeoff between available range and azimuth information can be successfully overcome. To achieve DoA estimation, the MUltiple SIgnal Classification (MUSIC) algorithm which is commonly used in phased-array antennas is extended to the novel imaging scheme based on FSRA, which is referred to as FS-MUSIC algorithm. The model of FSRA manifold vector is constructed for FS-MUSIC with a modified spatial smoothing method specifically developed to decorrelate the coherent back-scattering signals and to resolve the targets located within a same half-power beamwidth (HPBW). The resolution of the FS-MUSIC algorithm is derived based on the Bayesian approach. Superresolution imaging with real-time capability was demonstrated by both the simulation and the proof-of-principle experiments in the 0.3-THz band. An FSRA operating on 0.3-THz frequency band was developed.