Control System-Oriented MIMO Over-the-Air Computing

This paper applies the MIMO over-the-air computation (MIMO-AirComp) technique to compute control signal of a distributed system, including state estimation. The control-based target function is expressed as nomographic and efficiently computed through wireless channel with suitable decomposition. proposed scheme leverages local pre-processing at individual sensor, aggregation multiple access channel, post-processing controller control-oriented function. We analyze performance our in terms time delay error compared state-of-the-art time-division by means Monte Carlo simulations. Numerical results indicate that considerably improves performance, particularly noisy networks rely on large number sensors, low complexity for resource-constrained scenarios.