Latency and timeliness in multi-hop satellite networks

TL;DR

This paper analyzes delay and Age of Information in multi-hop satellite networks modeled as queue networks, revealing how system utilization and node unloading affect latency and timeliness for designing efficient satellite communication systems.

Contribution

It introduces a queue network model for analyzing delay and AoI in multi-hop satellite networks, providing new insights into optimizing latency and information freshness.

Findings

Minimum AoI occurs at lower system utilization as nodes increase.

Unloading initial nodes reduces queueing time and AoI.

Results inform design of latency-sensitive satellite networks.

Abstract

The classical definition of network delay has been recently augmented by the concept of information timeliness, or Age of Information (AoI). We analyze the network delay and the AoI in a multi-hop satellite network that relays status updates from satellite 1, receiving uplink traffic from ground devices, to satellite K, using K-2 intermediate satellite nodes. The last node, K, is the closest satellite with connectivity to a ground station. The satellite formation is modeled as a queue network of M/M/1 systems connected in series. The scenario is then generalized for the case in which all satellites receive uplink traffic from ground, and work at the same time as relays of the packets from the previous nodes. The results show that the minimum average AoI is experienced at a decreasing system utilization when the number of nodes is increased. Furthermore, unloading the first nodes of the chain reduces the queueing time and therefore the average AoI. These findings provide insights for designing multi-hop satellite networks for latency-sensitive applications.