Physics Informed Adaptive Antenna Control for High Speed Railway Links with Paired Monte Carlo Validation

Abstract Reliable railway connectivity depends on both the electromagnetic characteristics of the rooftop antenna and its response to changing propagation geometry. This paper integrates analytical antenna design with a trajectory-aware adaptive low-profile mobility controller (ALMC) and a paired numerical evaluation. Rectangular-patch synthesis, impedance matching, conductor loss and realized gain establish the physical basis for a low-profile radiator; slot loading is considered as a route to wider bandwidth. These relations motivate, but do not independently validate, the discrete gain, beamwidth and interference-rejection states used in a 3.6 GHz railway channel surrogate. Five controllers are compared over a 6 km route using 50 paired stochastic trials. Relative to fixed control, the risk-aware controller improves mean SINR by 3.989 dB, reduces outage by 13.009 percentage points and increases the fraction of route samples with at least 6 dB link margin by 12.160 percentage points. The corresponding approximate 95% confidence intervals are 3.934–4.044 dB, 12.297–13.720 percentage points and 11.449–12.871 percentage points. Greedy instantaneous SINR selection nevertheless achieves the best mean values of all five reported RF metrics. Recovered source settings and direct paired analysis show that risk-aware control also incurs 27.58 additional mode switches and 2.206 mJ more modeled control energy per route than greedy selection. All five controllers are included in the stress comparison. A leaky-feeder extension describes tunnel coverage without claiming validated handover performance. The contribution is an integrated physical rationale and controlled comparison of adaptive antenna policies, conditional on the assumed channel and antenna states; the predictor uses oracle look-ahead, and full-wave, hardware and railway field validation remain necessary. The paper has concentrated on upper bound analysis.