Honest note [CALC]: Friis free-space loss at 5.8 GHz over the true slant range + literature mid-points for body shadow (−2…−12 dB by pose, the single biggest uncertainty) + the BoxPro's four fused helix branches, each with its own geometry: conservative 10.7 dBic peak (estimators span 10.7–13.9), 37° beamwidth, parabolic roll-off, and nothing better than a −5 dBi sidelobe floor off-axis. The transmit side gets +1.6 dBic because the lying omni's doughnut maximum points at the DZ — that is the whole point of turning it 90°. Pose shadows are referenced to the sender on the helmet's chin mount (literature mid-points — not yet our measurement). Threshold: −105 dBm is receiver spec class (bench measurement D1 decides), −90 dBm is the worst-case floor. This extends build/rf from one beam to the real four — a calculation to be jumped, not a measurement.