Distributed Scheduling Algorithm For Optimizing Age Of Information In Wireless Networks

Age of Information(AoI)is a metric that describes the freshness of information from the perspective of the destination node.Whether in the Internet of Things or the Internet,the freshness of information can reflect the timeliness of information,which is very important for time-delay sensitive applications(such as autonomous driving,etc.).This thesis studies the AoI optimization problem in wireless sensor networks.The system consists of several wireless sensor nodes and a destination node to collect information.The wireless sensor nodes transmit the sensed information to the information collection node through a wireless channel.For any sensor node,its AoI refers to the time elapsed from the time when the node successfully sent the collected data to the information collection node last time to the current moment.Based on the above definition of AoI,the peak age of the sensor node can be defined as the AoI at the time when the data packet is successfully received,which can reflect the maximum AoI that can be reached at that time.However,due to the limitation of the wireless channel,the parallel transmission between multiple wireless sensor nodes will interfere with each other,which greatly reduces the timeliness of data transmission,and it is necessary to reasonably schedule the data transmission of wireless sensor nodes.On the other hand,the existing related work only studies the centralized scheduling strategy,which requires a central facility to coordinate wireless sensor nodes;however,this central infrastructure is not always available in some application scenarios.In this thesis,we propose a distributed scheduling algorithm to minimize the maximum peak age of sensor nodes.In this algorithm,each sensor node independently accesses the channel and transmits data according to the transmission probability,and adaptively adjusts the transmission probability through channel sensing.The algorithm does not require coordination of central facilities,nor does it require communication between wireless sensor nodes.It can achieve the goal of minimizing the maximum peak age by maximizing network throughput and ensuring fairness between sensor nodes.In addition,we also considered the influence of jamming on this distributed randomized scheduling algorithm.Because jamming will make the transmission probability of all nodes smaller,which indirectly affects the AoI of the system.The reason why jamming can affect the age of information is mainly because in the process of adjusting the transmission probability,we cannot distinguish whether the transmission failure is caused by jamming or the interference caused by the parallel transmission of sensor nodes,so we try to avoid the event of transmission failure has an impact on the adjustment process of transmission probability so as to avoid the impact of jamming on the age of information.Finally,we verified the performance and anti-jamming ability of the distributed scheduling algorithm through theoretical analysis and experimental simulation.