Optimizing element positioning in sparse arrays for nearfield mm-wave imaging

In order to minimize hardware and processing complexity in a nearfield mm-wave imaging, it is important to reduce the number of receiving elements as much as possible without significantly compromising reconstruction quality. While array thinning has been long practiced for farfield antenna systems, nearfield array thinning is novel in that the figure of merit (or cost function) is based on the wideband point spread function of the array when viewed as a transmitter. Simulated annealing is used to optimize the element positions for a quarter-circle array for 83% and 90% thinning with minimal and reasonable PSF distortion.