We present the design and simulation of a frequency-diverse aperture for imaging of human-size targets at microwave wavelengths. Predominantly relying on a frequency sweep to produce diverse radiation patterns, the frequency-diverse aperture provides a path to all-electronic operation, sampling a scene without the requirement for mechanical scanning or expensive active components. Similar to other computational imaging schemes, the frequency diverse aperture removes many hardware constraints by placing an increased burden on processing and analysis. While proof-of-concept simulations of scaled-down versions of the frequency-diverse imager and simple targets can be performed with relative ease, the end-to-end modeling of a full-size aperture capable of fully resolving human-size targets presents many challenges, particularly if parametric studies need to be performed during a design or optimization phase. Here, we show that an in-house developed simulation code can be adapted and parallelized for the rapid design and optimization of a full-size, frequency-diverse aperture. Using files of human models in stereolithography format, the software can model the entire imaging scenario in seconds, including mode generation and propagation, scattering from the human model, and measured backscatter. We illustrate the performance of several frequency-diverse aperture designs using images of human-scale targets reconstructed with various algorithms and compare with a conventional synthetic aperture radar approach. We demonstrate the potential of one aperture for threat object detection in security-screening applications.