Minimizing Age Violation Probability with Adaptive Blocklength Selection in Short Packet Transmissions.

As a measure of data freshness, Age of Information (AoI) has become an important metric in systems where small status update packets need to be delivered to a monitor in a timely manner. The goal of this study is to minimize the age violation probability, which is defined as the probability that instantaneous age exceeds a certain threshold. An adaptive policy using finite blocklength theory approximations and dynamic programming tools is proposed for choosing the optimal blocklength. The problem is defined as a Markov decision process (MDP), and a state-aggregated value iteration algorithm is adopted to find a policy that dynamically selects the optimal blocklength for the current state of the system. Performance of the proposed policy is demonstrated in different scenarios and compared with benchmark policies with a fixed blocklength.