Advancing Spectrum Sharing through Statistical Analysis of EESS-Passive Satellite Overpasses

Earth-observing measurements performed by Earth Exploration-Satellite Service (EESS) passive satellites are critical to meteorological and climatological assessments of the state of planetary conditions. Radiometers, which perform these measurements, are highly sensitive to anthropogenic transmissions. Therefore, other user access to the spectrum must be carefully managed during an overpass. In this work, we present a statistical characterization of interarrival times and durations of EESS-passive satellite overpasses to support studies between active and passive users co-existing within the radio spectrum. Our work analyzes datasets compiled from multiple EESS-passive satellites to determine the frequency and duration of service interruptions to support the design and evaluation of spectrum sharing approaches. EESS-passive satellite observations that fall within a 100~km region surrounding 15 selected cities are used to produce complementary cumulative density functions (CCDFs) for interarrival times and durations. While the CCDFs for the interruption durations appear location-independent, the interarrival CCDFs demonstrate a high correlation with city latitude due to the nature of EESS-passive orbits. The interarrival time mean and standard deviation are found to be similar and are characterized with simple quadratic functions of latitude that achieve R-squared values of 0.993 and 0.996. The similarity of the mean and standard deviation suggests that a Poisson arrival process (having exponentially distributed interarrival times) is a reasonable assumption in spectrum sharing studies, though more precisely fitted distributions are also discussed. We apply our findings to an economic model of spectrum sharing between EESS-passive and wireless communications services at a given location. We find that the location’s latitude has a significant impact on the profitability of the wireless provider.