Distribution of AoI in EH-powered Multi-source Systems with Source-aware Packet Management

This paper considers a multi-source updating system in which a transmitter powered by energy harvesting (EH) sends status updates about multiple sources of information to a destination, where the freshness of status updates is measured in terms of Age of Information (AoI). The harvested energy packets and the status updates of each source are assumed to arrive at the transmitter according to independent Poisson processes, and the service time of each status update is assumed to be exponentially distributed. Our focus is on understanding the distributional properties of AoI under a source-aware preemptive in service queueing discipline (which only allows preemption between the status updates generated by the same source to enhance fairness). In particular, we use the stochastic hybrid systems (SHS) framework to derive closed-form expressions of the moment generating function (MGF) and average of AoI. To the best of our knowledge, this paper is the first to characterize the AoI performance under a source-aware preemptive policy for the generic case where the transmitter has an arbitrary number of sources. The generality of our results is demonstrated by recovering several existing results for EH-powered single-source systems as special cases. Our results demonstrate that the proposed source-aware preemptive policy strikes a balance between minimizing the sum of average AoI values associated with different sources (average sum-AoI) and achieving fairness among the average AoI values of different sources.