ABSTRACT

The topic of bipolar junction transistors (BJTs) is obviously quite broad and a full treatment would consume volumes. This work focuses on basic principles to develop an intuitive feel for the transistor behavior and its application in contemporary high-speed integrated circuits (ICs). The exponential growth in high bandwidth wired, wireless, and fiber communication systems coupled with advanced IC technologies has created an interesting convergence of two disparate worlds: the microwave and analog domains. The traditional microwave IC consists of a few transistors in discrete form or in low levels of integration surrounded by a sea of transmission lines and passive components. The modern high-speed analog IC usually involves tens to thousands of transistors with few passive components. The analog designer finds it a more cost-effective solution to use extra transistors rather than passive components to resolve performance issues. The microwave designer speaks in terms of noise figure (NF), IP3, power gain, stability factor, voltage standing wave ratio (VSWR), and s-parameters while the analog designer prefers noise voltages, harmonic distortion, voltage gain, phase margin, and impedance levels. Present BJT IC technologies fall exactly in this divide and thus force the two worlds together. Since parasitics within an IC are significantly lower than those associated with packages and external interconnects, analog techniques can be applied well into the microwave region and traditional microwave techniques such as impedance matching are only necessary when interfacing with the external world where reference impedances are the rule. This symbiosis has evolved into what is now termed radio frequency IC (RFIC) design. With this in mind, both

analog and microwave aspects of bipolar transistors will be addressed. Throughout the discussion, the major differences between BJTs and field-effect transistors (FETs) will be mentioned.