MM-wave integration and combinations
Journal
IEEE Microwave Magazine
Journal Volume
13
Journal Issue
5
Pages
49-57
Date Issued
2012
Author(s)
Abstract
Millimeter wave frequency band (30300 GHz) radios are creating a strong demand for highly integrated transceiver monolithic microwave/ millimeter-wave integrated circuits (MMICs). Power amplifiers (PAs) are an important block in the wireless transceiver. PA designs often involve tradeoffs between available power gain, output power (Pout), power added efficiency (PAE), and linearity. Traditionally, millimeter-wave PAs were mostly implemented in compound semiconductor technologies such as GaAs and InP HEMT technologies due to their high electron mobility, high breakdown voltage, and the availability of high-quality-factor (Q) passives. On the other hand, SiGe BiCMOS technology has shown potential for medium output power at millimeter-wave frequencies. Modern nanometer CMOS technology with continuous downscaling of transistor dimensions and improvement of the maximum frequency of oscillation, fmax, has become a realistic alternative for millimeter-wave applications. Recently, CMOS MMIC components such as voltage control oscillators (VCOs), low noise amplifiers (LNAs), and mixers have been demonstrated successfully with good performance at millimeterwave frequencies above 100 GHz [1][3]. Due to the advantages of small size, low cost, low power consumption, and high level of integration with the back end, millimeterwave system-on-chip (SoC) technologies using nanometer CMOS processes have the opportunity to meet millimeter-wave system requirements; however, the CMOS PA is still a bottleneck for millimeter-wave system integration. Since Si has a lower mobility and smaller band-gap than III-V compound semiconductor materials, Si metal oxide semiconductor (MOS) field-effect transistors (FETs) have a lower cut-off frequency and breakdown voltage than compound semiconductor transistors such as GaAs-based metal-semiconductor FETs (MESFETs), HEMTs, and heterojunction bipolar transistors (HBTs). The downscaling of transistor dimensions, resulting in a low transistor breakdown voltage in nanometer CMOS is a disadvantage in millimeter-wave PA design. In addition, high Si substrate loss in commercial CMOS technologies is not conducive for power and efficiency of the PAs. Therefore, more circuit design effort has to be devoted to millimeter-wave PAs using nanometer CMOS technologies. © 2012 IEEE.
SDGs
Type
journal article