The Evolution of Microwave and MM-Wave CMOS ICs

April 14, 2015

Since the 1970s, researchers and engineers have been advancing the techniques and technology necessary to enable the next wave of low-cost and highly integrated RFICs.

Jean-Jacques DeLisle

Mwrf Com Sites Mwrf com Files Uploads 2015 02 05 E F1

1. Designing millimeter-wave, controlled-impedance interconnects requires detailed analysis of the material and geometric factors that are minor considerations at lower frequencies.

So far, demand for millimeter-wave electronics in commercial or consumer markets isn’t strong. However, if the cost structure for such devices became less restrictive, it may pave the way to countless applications. Millimeter-wave CMOS—with a high degree of integration at low costs—could be a potential enabler for these technologies. Postulating that different circuit approaches other than RF-CMOS are necessary to develop millimeter-wave CMOS, Asad A. Abidi of the Electrical Engineering Department of the University of California, Los Angeles, presented research on the progression of millimeter-wave CMOS development since the 1970s.

Since World War II, the military has shown interest in millimeter-wave technology for electronic warfare (EW). But reliable devices that could produce millimeter waves weren’t feasible until decades afterward. As part of a military grant, Texas Instruments developed silicon-based modules for phased-array radars in 1964. This spurred the development of hybrid integrated circuits (HICs), but they suffered from significant interconnect parasitics above 10 GHz. After years of further research, HICs gave way to monolithic microwave integrated circuits (MMICs).

Mwrf Com Sites Mwrf com Files Uploads 2015 02 05 E F2

2. Concerns over the propagation modes on substrate resistivity and frequency preside over devices such as phased-array systems-on-a-chip.

The first real MMIC—the hybrid semiconductor—exploited gallium arsenide (GaAs), with the justification that higher-electron-mobility devices would have a higher maximum operating frequency. Techniques for developing interstage matching elements between field-effect transistors (FETs) were first performed with GaAs ICs and later implemented in CMOS MMICs. Here, researchers carefully detailed transistor placement and use of low-inductance interconnects. This provided fertile ground for the advancement of silicon-based, millimeter-wave MMIC technologies.

Phased-array technologies with beam-steering capability continue to be a major application for millimeter-wave CMOS, driving the development of highly integrated millimeter-wave transistors. These applications may lead to reduced costs and the greater adoption of CMOS MMICs for radar, backhaul, 5G, and gigabit Wi-Fi. See “CMOS Microwave And Millimeter-Wave ICs: The Historical Background,” 2014 IEEE International Symposium on Radio-Frequency Integration Technology (RFIT), Aug. 2014, pg. 1-5.