Acceleration of wireless channel simulation using GPUs

High-fidelity site-specific modeling of wireless communications channels has historically been too computationally intensive for casual implementation in network simulators. However, simulation cannot predict the behavior of wireless networks in real-world environments without modeling the physical channel to the degree needed to support the objectives of the simulation. The most commonly used models for studies of networking protocols do not directly model domain geometry, and use statistical models of path-loss. However, simulations of wireless indoors, or out in a domain that is cluttered with obstacles need more resolution in the description of the domain. Realistic models for these cases typically involve large amounts of floating point computation, to which modern GPUs are well suited. In this paper we demonstrate parallel radio propagation prediction in a single machine using multiple GPUs and CPU cores. We explore the tradeoffs between model accuracy and performance, and use techniques from graphical raytracing to improve the speed with which radio path loss can be computed.