Skip to Main Content
A method is presented for deriving quasi-power invariants of linear or linearized networks in linear or linearized embedding in matrix and scalar form. These complement the adimensional "cross ratio" invariants discussed in an earlier paper. Connections and differences are illustrated by means of three practical examples: the Q-factor of a resonator, a generalized form of Hines' switching theorem, and a "figure of merit" for materials in an electromagnetic cavity. The wellknown noise matrix of Haus and Adler is recovered as a particular case of a more general form. A few new invariants are presented. The relationship with Tellegen's theorem (for the scalar case) is discussed.