Reading through the coverage test of the RAK7258 in the last 2 slides they indicate that by reversing the direction of the antenna from pointing upwards to downwards in their test, the loss % decreased for every node whether it was above or below the antenna.
Can someone make sense of why this is the case? If the gateway was placed on the 3rd floor, I could imagine the results from floor 4 and 5 swapping with 1 and 2, but it seems to have significantly improved results across the board.