In order to achieve the maximum power transfer, we need to match the impedance of the load to that of the source as already pointed out in Chapter 2. Usually this is accomplished by incorporating additional passive networks connected in-between source and load. These networks are usually referred to as matching networks. However, the functionality of matching networks is not simply limited to matching source and load impedances for optimal power flow. In fact, for many practical circuits matching networks are not only designed to meet the requirement of minimum power loss, but are also based on additional conditions such as minimizing the noise influence, maximizing power handling capabilities, and linearizing the frequency response. In a more general context, the purpose of a matching network can be defined as an impedance transformation to convert a given value to another, more suitable value.

In this Chapter we restrict our coverage to the techniques of performing impedance transformation using passive matching networks. The emphasis is to ensure minimum reflections between source and load. All other considerations such as noise figure and linearity are left for discussions in the subsequent Chapter 9.

We commence with a study of networks based on discrete components. These networks are easy to analyze and can in practice be used up to frequencies in the low GHz range. Next, we will continue with the analysis and design of matching networks using distributed elements, such as striplines and stub sections. These networks are more suitable for operational frequencies exceeding 1 GHz, or for cases where vertical circuit dimensions are of importance as required in RF integrated circuit designs. In order to simplify our treatment and to gain clarity in the design methodology, the Smith Chart as a primary design tool will be utilized extensively throughout.