The high frequencies utilized by automotive radar sensors, coupled with the electrically large fascia in front of the sensors, have posed a challenge for simulation software in the form of long simulation times. Advances in NVIDIA graphics processing units (GPUs) have alleviated this problem, increasing the productivity of RF engineers who need high fidelity simulations of a sensor behind fascia. Using a 25 GHz short-range radar (SRR) model, this paper compares multiple GPU architectures released in 2007 through 2013 to look at the downward trend in simulation times. A larger than 65% decrease in simulation times is observed, which allows a fully detailed sensor simulation to run in less than 4 hours on modern GPUs.

Determining an optimal position of an antenna on a platform is not always intuitive. This paper examines the implementation of a Genetic Algorithm (GA) within XFdtd, a Finite Difference Time Domain (FDTD) solver, for its effectiveness in determining an optimal antenna position when compared to a brute force method.XFdtd uses the FDTD method [1] to solve general electromagnetic problems. Built into XFdtd is the concept of Feature Based Modeling (FBM), which uses relative coordinate systems to position antennas in a geometry in relation to other parts. The GA, written in C++, is available to XFdtd's Scripting API. Through scripting a user specifies the antenna to move and provides bounded regions where the antenna can be located for maximum efficiency.

XFdtd's GA implementation successfully found optimal solutions for the two models used in this paper. Defining the antenna to optimize and the acceptable regions where the antenna can be located was done through the Scripting API. Also, the cost was determined in a script function and passed to the GA for use in natural selection.