Timely Status Update: Should ARQ be Used in Two-Hop Networks?

TL;DR

This paper analyzes the impact of ARQ on the freshness of information in two-hop networks, revealing that ARQ can improve average AoI, unlike in single-hop networks, with theoretical and simulation validation.

Contribution

It provides a theoretical analysis of average AoI in two-hop networks with ARQ, showing its benefits and the importance of second hop success probability.

Findings

ARQ reduces average AoI in two-hop networks.

Second hop success probability significantly affects AoI.

ARQ's benefit is not observed in single-hop networks.

Abstract

This paper investigates the information freshness of two-hop networks. Age of information (AoI) is used as the metric to characterize the information freshness, defined as the time elapsed since the latest received status update was generated. In error-prone wireless networks, prior studies indicated that Automatic Repeat-reQuest (ARQ) does not help improve the average AoI performance of single-hop networks, because sending a new packet always carries the most up-to-date information (i.e., discarding the old packet). We believe that this observation does not apply to two-hop networks. For example, when a packet transmission fails in the second hop, although a new packet has more recent information, it may require more time to be delivered (i.e., the communication has to restart from the first hop), thus leading to a high AoI. This paper analyzes the theoretical average AoI of two-hop networks with and without ARQ. Specifically, we model the two schemes using Markov chains, from which we derive the average AoI. Our theoretical and simulation results confirm that, unlike single-hop networks, ARQ should be used in two-hop networks to achieve lower average AoI. In particular, when ARQ is used, the successful decoding probability of the second hop has a greater impact on the average AoI than that of the first hop. Overall, our findings provide insight into the ARQ design for two-hop timely status update systems.