Keynote speaker 2: Combining computational electromagnetics with signal processing algorithms to enhance the performance of imaging devices and antennas

Computational Electromagnetics (CEM) has made great strides over the years and has enabled us to solve very large and complex problems that were well beyond our reach only a few year ago. This has been made possible because of steady increases in computing speeds - thanks to Moor's Law - and the advent of parallelization as well as development of iterative and domain decomposition techniques. The quest as well as progress toward the solution of very large and complex problems remain unabated - CEM problems involving "billions and billions" of unknowns are being routinely solved today - and this trend is likely to continue for a long time to come. The focus of this presentation is not on proposing yet another algorithm which would help us solve even larger problems than we can handle today, but on discussing alternate ways by which we can enhance the performance of microwave devices - such as imaging systems and lenses - by combining signal processing algorithms with CEM. The paper will discuss several examples of performance enhancement including sub-wavelength imaging of objects that are not located in the near fields of lenses, and improving the accuracy of direction finding antenna systems with size constraints that limit their limitation if conventional DF techniques are used. Our strategy is to first use available CEM techniques to solve the forward problems efficiently and then to use these solutions in Signal Processing algorithms for the purpose of performance enhancement of microwave devices of the type mentioned above.