Design of an on-chip Hilbert fractal inductor using an improved feed forward neural network for Si RFICs

Abstract

This paper presents an efficient modeling of Hilbert fractal inductors by improved feed forward neural network trained hybrid particle swarm optimization and gravitational search algorithm (FNNPSOGSA). The proposed model computes the effective inductance value (L) and quality factor (Q) of Hilbert fractal inductors with metal trace width, effective fractal length, frequency, and oxide thickness as input parameters. In contrast to the traditional feed forward neural network, the proposed FNNPSOGSA has been designed with fewer hidden neurons with much-enhanced learning and generalization capabilities. As a consequence, the proposed model achieves better speed and is as accurate as electromagnetic simulations. From the simulation results, it is proved that the proposed model is a good alternative for complex fractal inductor design.

A novel hybrid series stacked differential fractal inductor for MMIC applications

“…Inductors making use of fractal geometry acts as the precise technique to resolve this problem. An exhaustive study of inductors implemented using fractal curves was carried out (Lazarus et al, 2014;Kumar and Rao, 2015;Maric et al, 2008;Padavala and Nistala, 2018;. The performance of the fractal inductor was further enhanced by using fractal loop structures (Shoute and Barlage, 2015;Padavala and Nistala, 2017).…”