Information freshness optimization of multiple status update streams in Internet of things: Generation rate control and service rate reservation

Abstract

The Internet of things (IoT) has become a key infrastructure providing up-to-date and fresh information for policy analysis and decision-making of upper-layer applications. However, there are limited sensing and communication resources in IoT devices, which significantly affects the timeliness and freshness of the updated status. This work proposes two schemes, namely, the generation rate control and service rate reservation schemes, to improve the overall information freshness of multiple status update streams at the receiver. Specifically, using the recently proposed Age of Information (AoI) as the metric for evaluating information freshness, we characterized the overall information freshness, i.e., the overall average AoI at the receiver for both schemes, by considering the urgency difference of status update and streams. Both schemes for status updates and streams, respectively, were formulated as two optimization problems. We proved that both problems are convex and the optimal generation and service rates for different streams are found by the standard convex optimization algorithm. Moreover, we proposed both approximate optimal generation and approximate optimal service rate for fast deployment in heavy and light load cases. Numerical results verify the theoretical findings and accuracy of the proposed approximate solutions, guiding the design and deployment of IoT.