Abstract

This paper utilizes a novel approach to analyzing remote monitoring systems by using the Age of Information (AoI) as a metric to measure the timeliness of status updates. The study focuses on a scenario with a single source and destination queueing system, two parallel heterogeneous servers, and no additional buffer, where the analysis of AoI or PAoI is difficult because of the out-of-order reception of packets. To analyze the performance of the system, we use stochastic hybrid systems (SHS) approach. We evaluate the AoI and Peak AoI (PAoI) of the system using different queueing disciplines, including Last Come First Serve (LCFS) and LCFS with probabilistic routing. We use Proactive Obsolete Packet Management (POPMAN) method to identify obsolete packets that can be discarded in advance, which saves server processing time and receive packets in order at the receiver. The study compares the performance of this system to a similar system with two homogeneous servers. The results show that by using the POPMAN method, there is an improvement in the AoI and PAoI for LCFS queueing system and the performance of LCFS with probabilistic routing using the POPMAN approach is same as the traditional scheme. Additionally, the study analyzes various parameters, such as the probability of preempted packets, the probability of obsolete packets, the probability that a packet is informative or successfully delivered, and optimal splitting probabilities for probabilistic routing.