Precoding Design for Ensuring Data Freshness in Multi-User MISO Networks.

The existing multiple-input single-output (MISO) design to transmit data reliably to multiple mobile users (MUs) has become insufficient as MUs require fresh data, not just data. Therefore, in this paper, we consider multi-user MISO networks, where a base station (BS) equipped with multiple antennas serves MUs that want to maintain data freshness. To analyze the data freshness in this network, we first define the age-of-information (AoI) violation ratio, which is the ratio of duration that the AoI is larger than the AoI violation threshold to a sensing period. We then obtain the upper bound on the AoI violation ratio using the smooth maximum/minimum unit. We consider an optimization problem that minimizes the AoI violation ratio of multiple MUs. We propose a generalized power iteration (GPI) precoding algorithm to find a principal precoding vector that satisfies a first-order optimality condition of the optimization problem. Furthermore, for the scenario where the BS has the imperfect channel state information (CSI) of MUs, we provide the upper bound on the AoI violation time using the lower bound on the ergodic spectral efficiency, and also design the GPI precoding algorithm. Simulation results show proposed methods outperform baseline methods and demonstrate the effect of network parameters on the AoI violation probability.