Enhancement of a gain-sensitivity analysis of electrical networks is presented by computing gain sensitivities with respect to network parameters. A simple and versatile method. The so-called chain-sensitivity matrix is presented and compared with the current method in the literature, gain factorization, for the gain sensitivities of the cascaded networks. Analytical formulas are derived to calculate gain sensitivities of the T and Pi types of distributed-parameter amplifiers with respect to the physical length l and characteristic impedance Z(0), rather than using a time-intensive computer-based perturbation method. The numerical results of the T- and Pi-type amplifiers for the design targets of noise figure F-req = 0,46 dB (double left right arrow 1, 12) input VSWR V-ireq = 1, power gain G(Treq) = 12 dB (double left right arrow 15, 86) and the bandwidth B = 2 GHz - 11 GRZ are given in comparison to each other. (C) 2004 Wiley Periodicals, Inc.