We rely on radio frequency or RF systems for all kinds of everyday devices, from cell phones and wireless routers to televisions, remote controls and set-top boxes. They can even be found in the more advanced medical devices that are implanted to help cure pain and illness.
Researchers at Aalborg University (AAU) in Denmark are applying NVIDIA GPUs and the AccelerEyes Jacket software platform for MATLAB to these systems to advance their theory, design and implementation.
Dr. Torben Larsen’s team at AAU is using NVIDIA CUDA and AccelerEyes technologies to not only develop new algorithms for these systems more rapidly but to run those new algorithms in a fraction of the time that it took on a CPU. Data interpolation and Fast Fourier Transforms (FFT) for example, are common tasks performed in signal processing applications. Dr. Larsen’s team found these tasks run 25 to 35 times faster with Jacket and NVIDIA GPUs over CPUs.
The hope is that by increasing the speed that these systems can be developed and tested, they can bring more effective and safer RF-enabled devices to market. More information on the work being done by Dr. Larsen’s team and the impact that GPU computing had had on his research can be found in this recent story. Also, be sure to visit Torben’s corner for tips and tricks on how to maximize the use of GPU technology from NVIDIA and AccelerEyes.