An approach for knowledge-aided-design (KAD) of microwave circuits using artificial neural networks

This paper points out that appropriately trained artificial neural networks can be used to develop knowledge-aided-design (KAD) modules for RF and microwave design. The approach is illustrated by examples of the reflection type and the loaded line type phase shifters. It also is applicable to other types of circuits and subsystems. KAD modules were developed and verified by using available microwave CAD software.