Statistics of correlated inverse-gamma random-variable sums

Investigate the corresponding statistics of sums of correlated inverse-gamma random variables, extending the independent, not necessarily identically distributed inverse-gamma framework to settings with spatially correlated shadowing in UAV-assisted communications.

Background

The paper derives exact PDF and CDF expressions for sums of independent, not necessarily identically distributed inverse-gamma random variables, under the restriction that the shaping parameters are integer plus one-half. These results are applied to outage and coverage analysis in UAV-assisted wireless networks and high-speed railway communications.

The conclusion identifies correlated inverse-gamma sums as an unresolved direction. Such an extension is motivated by UAV-assisted communication environments, where shadowing may be spatially correlated across different UAVs and along successive waypoints of a single UAV trajectory.