Peak Age of Information Optimization in Heterogeneous Aloha Networks

Rapid advancements in Internet of Things (IoT) technology have promoted numerous novel applications that are sensitive to information timeliness. With different applications coexisting in one wireless network, each of which has heterogeneous traffic characteristics, it is of paramount importance while challenging to address how to characterize and optimize the information freshness performance network-wide, particularly in the massive access scenario. This paper casts attention on the heterogeneous slotted Aloha network and uses the Peak Age of Information (PAoI) metric to quantify information freshness. By assuming that the sensors in each group are equipped with unit-sized buffers and Bernoulli packet arrival, we derive the network steady-state point and PAoI. Depending on whether one group or all groups are age-sensitive, we focus on single-group PAoI and global PAoI optimization respectively. When only one single group is age-sensitive, we derive its optimal transmission probability. When all groups are age-sensitive, we propose a heuristic algorithm based on the particle swarm optimization method, where the bi-stability of Aloha network is considered for avoiding the risk of rapid performance deterioration. Extensive simulation results are presented to verify our analysis and the effectiveness of the proposed algorithm.