Terahertz radar based on leaky-wave coherence tomography can detect the location of an object by finding the peak of the acquired frequency spectrum and its inverse Fourier transform, each corresponding to the angle and range of the object, respectively. While this approach is promising to implement high-resolution and compact radar, we find that locations of multiple objects cannot be determined uniquely. To avoid this ambiguity, here we propose stepwise signal processing. We experimentally demonstrate the unambiguous detection of two identical objects located at different points.
This paper reports on our industry-academia project of using a functional language in business software production. The general motivation behind the project is our ultimate goal of adopting an ML-style higher-order typed functional language in a wide range of ordinary software development in industry. To probe the feasibility and identify various practical problems and needs, we have conducted a 15 month pilot project for developing an enterprise resource planning (ERP) system in SML#. The project has successfully completed as we have planned, demonstrating the feasibility of SML#. In particular, seamless integration of SQL and direct C language interface are shown to be useful in reliable and efficient development of a data intensive business application. During the program development, we have found several useful functional programming patterns and a number of possible extensions of an ML-style language with records. This paper reports on the project details and the lessons learned from the project.