Comparing different approaches for stellar intensity interferometry

Stellar intensity interferometers correlate photons within their coherence time and could overcome the baseline limitations of existing amplitude interferometers. Intensity interferometers do not rely on phase coherence of the optical elements and thus function without high-grade optics and light combining delay lines. However, the coherence time of starlight observed with realistic optical filter bandwidths (> 0.1 nm) is usually much smaller than the time resolution of the detection system (> 10 ps), resulting in a greatly reduced correlation signal. Reaching high signal-to-noise ratio in a reasonably short measurement time can be achieved in different ways: either by increasing the time resolution, which increases the correlation signal height, or by increasing the photon rate, which decreases statistical uncertainties of the measurement. We present laboratory measurements employing both approaches and directly compare them in terms of signal-to-noise ratio. A high-time-resolution interferometry setup designed for small-to-intermediate-sized optical telescopes and thus lower photon rates (diameters < some metres) is compared to a setup capable of measuring high photon rates, which is planned to be installed at Cherenkov telescopes with dish diameters of > 10 m. We use a xenon lamp as a common light source simulating starlight. Both setups measure the expected correlation signal and work at the expected shot-noise limit of statistical uncertainties for measurement times between 10 min and 23 h. We discuss the quantitative differences in the measurement results and give an overview of suitable operation regimes for each of the interferometer concepts.