This award will support collaborative research between Prof. Herbert Carlin of Cornell University and Prof. Pier P. Civalleri of the Istituto Politecnico di Torino in Turin, Italy. The design of broad band analog circuits, particularly at frequencies in the microwave spectrum and higher, has become increasingly important in recent years. The application to very broad band microwave field effect transistor (FET) amplifiers is of great current interest. The theoretical basis for such applications is gain-bandwidth theory or, equivalently, broad band matching theory. Previous work by the U.S. P.I. and others has led to the development of a new bandwidth theory, called the real frequency method, as an alternative to analytic gain-bandwidth theory. The latter, dependent on analytical specifity of the appropriate function of the complex frequency variable, has several drawbacks that make it numerically intractable for loads of modest complexity. By contrast, the real frequency method handles loads of arbitrary complexity and directly utilizes real frequency numerical data. The investigators plan to extend the real frequency method they have developed for single input-single output systems with complex terminations, to the n-port equalizer problem. They also plan to specifically concentrate on viewing a multi-element distributed amplifier as a multichannel problem so as to develop simple parameter structures with optimum flat gain response. The investigators bring to the collaboration past successful experience in working jointly on research problems in this area. Wide band multichannel equalizers find important applications to systems in which multiplicity of complex loads must be simultaneously matched to one or more sources. Examples are generalized frequency crossover networks, multiplexing applications, and broad band amplifiers using many active elements.
