A maximally informative axion haloscope analysis

Axion haloscopes attempt to infer the presence or absence of dark matter axion-induced fluctuations in the electromagnetic field. The statistical analysis of the haloscope data should make as efficient use as possible of its information content, while remaining robust against bias and practical non-idealities of the measurement. We present a Bayesian analysis framework that exploits all of the information available from haloscope search data and allows freedom to alter scan procedures in light of measured data. Performing a reanalysis of the Phase I data from the HAYSTAC experiment, we more tightly constrain the dark matter axion parameter space, achieving the equivalent of a 39% speedup in the axion scan rate. By switching from the presently used threshold-based analysis framework to one that continuously monitors the measured power, haloscopes can achieve significant improvements in measurement ease and speed for no additional hardware or operating cost. I will discuss the details of the analysis technique and the implications for HAYSTAC and other haloscope-based axion dark matter searches.