Digital Over-the-Air Computation with Space-Time Processing for 6G Massive IoT Networks

In this paper, we propose a novel digital over-the-air computation (AirComp) framework employing space–time line codes (STLC) to achieve optimal spatial diversity gain with only local channel state information (CSI) at transmitters in uplink scenarios. The proposed technique alleviates the overhead of conventional AirComp, where the access point (AP) must acquire global CSI from all sensors, and is tailored to compute target functions under peak power constraints. Each sensor uniformly quantizes its measurement, applies bit slicing with low-order modulation, and transmits STLC-encoded symbols over two time slots. The AP subsequently performs linear combining of the received signals to detect the superimposed symbols and reconstruct the target sum. Simulation results demonstrate that the proposed technique achieves lower normalized mean-squared error than conventional analog and digital AirComp approaches.