Optical upconversion techniques for high-sensitivity millimetre-wave detection

Millimeter-wave radiation has the unique ability to penetrate atmospheric obscurations such as smoke, fog, and light rain while maintaining the capability for high-resolution imaging. However, suitable technologies for creating high-sensitivity, large pixel-count detectors are a limiting factor in the implementation of such systems. To this end, we present a technique for detecting millimeter-wave radiation based on optical upconversion that promises both high sensitivity and scalability to large pixel arrays. High-speed optical modulation is used to transfer millimeter-wave radiation onto the sidebands of a near-infrared optical carrier frequency. Optical filtering techniques are subsequently used to suppress light at the carrier frequency. The resultant signal is passed to a low-frequency photodetector, which converts the remaining sideband energy to a photocurrent proportional to the incident millimeter wave energy at the modulator input. Utilizing the low noise powers of such photodetectors, high sensitivities may be obtained even accounting for the relatively high signal losses associated with optical upconversion. Since optical upconversion inherently preserves both phase and amplitude information and fiber optics may readily be used for low-loss routing of the modulated signal, such an approach offers promise for high-resolution synthetic aperture imaging. Alternatively, since each of the required components may be fabricated in III-V materials using planar semiconductor processing techniques, integration of multi-pixel arrays is feasible. Herein, we present experimental results obtained using a baseline detector assembled from commercially available fiber-optic components as well as efforts to integrate the desired functionality into a single GaAs substrate. An initial noise equivalent power (NEP) of the proposed detector has been demonstrated at sub-nanowatt levels, with improvements to sub-picowatt NEP's anticipated as the setup is optimized.