Research: Novel Architectures for High-Frequency Electronics
The 21st century will usher unprecedented growth in the exchange of information. Telecommunication networks will support an array of services including telephony, fax, interactive video, mobile commuting, paging, and wireless Internet. In one way or another, all of these technologies will depend on high-frequency electronics, so the need for engineers trained in this field is great.
There is no question that high-frequency electronics will play an increasingly important role in electrical engineering education. This decade has already witnessed a more-than-tenfold increase in the clock speed of personal computers, so engineering students specializing in digital electronics must also be aware of such topics as electromagnetic interference and transmission-line coupling. The explosive growth in high-speed wireless communication systems, some of which have already migrated into the millimeter-wave range, will create the need for engineers who can develop flexible new circuit architectures, such as active antenna arrays.
Quasi-Optical Electronics
In contrast to conventional electronic circuits in which the input and output signals are voltages and currents, quasi-optical circuits have input and output signals that are electromagnetic beams. Quasi-optical circuits use components typically associated with optics, such as lenses, mirrors, and polarizers, but are targeted toward the millimeter and sub-millimeter-wave regimes of the electromagnetic spectrum between 30 GHz and a few THz. Research at UH focuses on active quasi-optical circuits used for generating, amplifying, and processing millimeter-wave beams.
Current research efforts are directed toward feedback optimization of grid oscillators, elimination of surface waves in active grids, and the design of new omnidirectional active arrays.
More about quasi-optics
Integration of Active Devices with Photonic-Bandgap Structures
A photonic crystal (PC) is an artificial dielectric structure whose electromagnetic dispersion relation has a band structure similar to that for electrons in crystalline solids. Electromagnetic waves incident on the PC can interfere constructively or destructively, creating stopbands or passbands. Current research efforts are focused on integrating PCs with solid-state devices, and developing novel planar photonic-bandgap structures.
Linearization of High-Efficiency Amplifiers
Analog and digital communication systems alike require linearity in the transmitter. Nonlinearity results in intermodulation distortion for analog systems and adjacent channel interference for digital systems. For modulation schemes such as quadrature amplitude modulation (QAM), in which information is encoded in both the amplitude and phase of the carrier wave, it is clear that both amplitude and phase linearity are needed.
Efficiency is another important issue in communication transmitters, particularly for mobile systems that depend on battery power. By far, the majority of the DC power in an RF transmitter is consumed by the power amplifier. To prolong battery lifetime, the power amplifier is typically driven into saturation to meet system requirements for high efficiency and high power. However, the saturation regime where high efficiency occurs is also the regime where linearity degrades. High power and efficiency performance therefore comes at the expense of distortion in the transmitted signal. This degradation comes in the form of both amplitude and phase distortion, since the gain and phase shift of the amplifier are functions of the input drive level. In situations where linearity is paramount, the amplifier must be backed off, sacrificing efficiency and output power.
One solution to this dilemma is to insert a predistortion linearizer having an amplitude and phase nonlinearity that is the inverse of the amplifier's into the signal path immediately preceding the amplifier. The result is a linear output signal over a greater dynamic range.
Current research is focused on new linearizer architectures specifically targeted for high-efficiency millimeter-wave power amplifiers.
Integrating Research and Teaching
I've developed an integrated research and education plan (spanning high school through post-graduate) to strengthen the high-frequency electronics skills of our nation's future electrical engineers. I'd be happy to discuss it if you swing by my office. This plan is essentially an inquiry-based approach for allowing undergraduates to actively participate in an integrated research-teaching environment, using high-frequency electronics education as a vehicle. My philosophy is based on educator John Dewey's observation almost a century ago: that true learning is based on discovery guided by mentoring rather than on the simple transmission of information.
The plan is hierarchical in approach, in that both undergraduate and graduate students serve as mentors for their younger counterparts in carefully tailored research projects. As undergraduates advance through the various programs, their learning experiences evolve closer to those of a graduate student. By the senior year, the undergraduate should be experienced enough to carry out a capstone design project which is approximately as complex as that of a first-year graduate student's.
A number of undergraduate students are working on cutting-edge research projects in my group. Several of them, denoted by the (*), were selected as NASA Space Grant Fellows and Trainees, collectively receiving over $40,000 to carry out their research.
Graduate students act as mentors for their younger counterparts, and in the process learn valuable management and training skills:
Positions available for Graduate Research Assistants and Post Docs
Opportunities are available for performing research in a funded research program integrating the areas of quasi-optical electronics, photonic-bandgap structures, and microelectromechanical systems. Please contact me for more information.
Last Modified: 9/23/00