LAUSR

We study a problem of scheduling transmissions of status messages over a fading link. The fading process is i.i.d., and is known at the sender after one-slot delay. Messages arrive according to an i.i.d. Bernoulli process at the sender, and each message has a delay deadline of ${d}$ time-slots. The problem is to decide whether or not to transmit, and to choose an optimum transmit-energy, such that the average Age of Information is minimum satisfying a constraint on transmit-energy and throughput. We pose this problem as a constrained Markov decision process, and provide an Age OPTimum policy (AOPT). We also propose a simple heuristic, Greedy Randomization of Energy Policy (GREP), and compare the performance of AOPT and GREP with ARQ policies. This is one of the first work in finding a minimum age scheduling policy with throughput guarantees and with delayed CSIT for an average transmit-energy constraint.