Hardware accelerated computing for electromagnetics applications

The current antenna technology is driven by both military and commercial applications to achieve multi-functionality with persistent connectivity in an integrated platform. The packaging and performance of the antenna in this integrated platform are thus critical factors to consider as an antenna designer. The need for reliable and efficient numerical techniques has been growing as the designs get more complex and a good prediction of system performance becomes essential for cost reduction. There are numerous ways of addressing this issue, such as developing hybrid methods that can avoid the numerical inefficiency of full wave methods while comparable accuracy can be achieved as effectively by asymptotic techniques. Another approach is to implement fast computational methods that utilize parallel computing platforms. It is the latter that is the focus of this paper with a particular focus on the use of general purpose graphics processing units (GPGPU) and field programmable gate arrays (FPGA). This paper will investigate both of these platforms in their applications to numerically intensive electromagnetic simulations. Weaknesses and strengths of both platforms will be investigated in the context of ease of use, efficiency, and potential for accelerated computations.