Makes sense that with more data about the physical orientation and characteristics of the device, a better signal attenuation attained and the network can be optimized. However, I don't see the logics of actually implementing such a scheme will be overcame, and ultimately this would have to become a part of the core communications protocol with the tower.