Contents I. Introduction
Currently, Age of Information (AoI), which can measure the freshness of data [1], [2], has drawn increasing attention since it was first proposed in [3], [4]. Formally, it is defined as the interval between the current time and the time when the most recent update (stored at the edge or cloud) is generated. Compared to the traditional packet-based metrics, such as latency and throughput, it can naturally characterize the freshness of data from the destination's perspective. Considering a typical scenario at network edge, which consists of a Base Station (BS) and multiple source nodes targeting for different applications. When the BS tries to collect the time-sensitive data from these source nodes, due to the bandwidth or energy constraints, only a part of source nodes can be scheduled to transmit their sampled data per time slot. Since the AoI values of these source nodes would grow linearly when no new sampled data are received at the BS, thus, how to obtain the scheduling strategy of data transmission to reduce the average and peak AoI has become a hotspot. Many efficient methods have been proposed by focusing on the scheduling of data transmissions [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25].
