Link Resource Allocation Strategy Based on Age of Information and Sample Extrusion Awareness in Dynamic Channels

Abstract

As an emerging measure of data freshness, the age of information (AoI) is receiving extensive attention. Many methods using AoI have been proposed for IoT communication scheduling. However, most of them are aimed at constant channel conditions in the ideal state, and the utilization of link resources is not sufficient. In addition, only the optimization of AoI is considered, without considering whether the sample is extruded or not. The occurrence of sample extrusion means that the transmission of the remaining untransmitted sample of the source node cannot be completed within the transmission time interval (TTI) before the arrival of the new sampling period of the source node, resulting in a phenomenon in which the new sample arrives while the old sample has not yet been completely transmitted. This scenario has a serious impact on delay-sensitive IoT applications. Therefore, under dynamic channel conditions and limited link resources, this paper establishes a mathematical model for AoI and sample extrusion. The influence of the scheduling algorithm on these two target values is analyzed and proved. Based on a greedy strategy, we propose an online allocation algorithm of preemptive link resources that considers two objectives: to give full play to the value of link resources and to minimize sample extrusion. The simulation results show that the proposed strategy can achieve better comprehensive performance in two scenarios where the sample variance between each source node is small and large.INDEX TERMS Age of information, sample extrusion, dynamic channel, preemptive.