Average Age of Information in Multi-Source Self-Preemptive Status Update Systems with Packet Delivery Errors

This paper studies the "age of information" (AoI) in a multi-source status update system where multiple sources send updates of their process to a monitor through a last-come first-served server with preemption in service and packet delivery errors. Arrival times of the status updates from the sources are assumed to be random according to independent Poisson processes. Service times are also assumed to be exponentially distributed and independent of the status arrivals. If the server is idle, any arriving packet immediately enters service. When the server is busy, if the arriving packet and the packet in service are from the same source, the packet in service is preempted and the new packet immediately enters service. Otherwise, any arriving packet is discarded. A closed-form expression for the average AoI of each source as a function of the system parameters is derived and, for the case without packet delivery errors, is compared to the average AoI in the "source agnostic" preemption setting considered by Yates and Kaul where any source can preempt any other source. The results show that source agnostic preemption in service results in better average AoI than self preemption in service for all sources.