An Automatic Model Order Reduction of a Uwb Antenna System

Abstract

Abstract-This paper proposes an efficient and automatic means of achieving a reduced model of a transfer function for UWB antenna design. According to the formulation of a transfer function, we have derived two factors, which are critical in determining the radiation pattern and input impedance respectively. Their special formula allow us to establish a reduced model automatically using the Model Order Reduction (MOR) techniques of a second order system. The process is free of any human factors and suitable to any antenna systems, thus enabling a direct and efficient interface with the optimization process in the design of a UWB antenna system. In addition, the proposed way of establishing a transfer function of the whole antenna system has successfully cascaded the entire system into separate subsystems, thus offering deeper insights in analyzing a UWB antenna system.

A Robust Cad Tool for Integrated Design of Uwb Antenna System

“…As verified in [26], this method is particularly suitable for the optimized design of a UWB antenna system. Firstly, it is highly efficient, thus relieving the computation pressure of optimization typically involving a large number of iterations.…”

“…For the details and verifications of this method, please refer to [26]. Although developed in the context of MoM, the method is applicable to any computational method or software that is formulated by the electrical integral function.…”

“…As a major part of this tool, we incorporate the method proposed in [26] to facilitate the characterization of a UWB antenna system. As verified in [26], this method is particularly suitable for the optimized design of a UWB antenna system.…”

“…While optimization becomes an indispensable part in present antenna design [21][22][23][24][25], the modeling, however, is expected to automatically proceed, for the need of working with a purely human-free optimization code. In our previous work [26], a new approach has been proposed, which automatically renders an analytical model of transfer function without any human input.…”

“…Firstly, in order to measure the extent to which the antenna system fulfills the given goals, the objective functions are redefined and established in terms of the transfer functions. For the convenience of incorporating with the optimization code, the approach proposed in [26] is applied to efficiently generate an analytical model of a transfer function. In the optimization block, a multi-objective algorithm, NSGA II [29] is adopted as the optimization solver.…”