CU-Simulator: A Parallel Scalable Simulation Platform for Radio Channel in Wireless Sensor Networks

Due to the computational intensive nature, the current available WSN simulators, which are based on the traditional CPU computing architecture, cannot run in a linear scalability. In this paper, we propose and set up CU-Simulator, a parallel radio channel simulator to enhance the performance for simulating data packet transmission in WSNs using NVIDIA’s CUDA-enabled GPU parallel computing architecture. First, the node positions are simulated on GPU. Second, we propose an efficient data structure for acceleration, called CUDA-quad-trees, residing in the fast on-chip memory of GPU, to organize sensor nodes in such a manner that the detection of possible transmitters is facilitated. Third, a CUDA parallel radio channel simulating engine is established. Experimental results show that CU-Simulator has a super-linear scalability and greatly outperforms a CPU implementation with up to 452.07-times speedup on an HP Z800 workstation with a NVIDIA Tesla C2070 card and an Intel Xeon Core-quad CPU.